Coding for animal ID marking

Abstract

Symbologies suitable for use marking codes on rodent tails are disclosed, for both machine-readable and human-readable codes, and those codes combined. Codes comprising both a human-readable and a machine-readable portion may associate with an animal ID within one study, and a study ID within a vivarium, respectively. A human-readable code uses a subset of an alphanumeric symbol set. A machine-readable code uses a vine-code with a visible spine aligned with the rodent's tail. Both symbologies support a system for deleting symbols and correcting codes marked in error. Both symbologies use symbol sets comprising only horizontal and vertical line segments. Both symbologies include a sparse mapping table for mapping marking codes to animal and study ID, and for error detection and for code re-use.

Claims

1. A marking code for encoding information on a rodent's tail comprising: a machine-readable code portion comprising a plurality of evenly spaced bars of varying length perpendicular to the tail of the rodent; a human-readable code portion comprising a plurality of human-readable alphanumeric characters; a first data table; a second data table; a first predetermined map that translates from the machine-readable code portion to a first animal identifier; a second predetermined map that translates from the human-readable code portion to a second animal identifier, distinct from the first animal identifier; wherein the first data table contains an entry comprising a first study value at a table index position of the machine-readable animal identifier; wherein the second data table contains an entry comprising a second study value at a table index position of the human-readable animal identifier; wherein the machine-readable code portion and the human-readable code portion are visibly marked on a rodent tail.

2. The marking code of claim 1 wherein: the machine-readable code portion and the human-readable code portion do not code the same data.

3. The marking code of claim 1 wherein: the machine-readable code portion and the human-readable code portion do not code the same data and the data coded by each is not a subset of the other data.

4. The marking code of claim 1 wherein: the entries in the second data table for valid second study values are located at indexed table lines wherein the index equals the valid second study value; and wherein at least one table line other than the above comprises an error indicator.

5. The marking code of claim 1 wherein: the second animal identifier is unique within a set of all animals in a single animal study.

6. The marking code of claim 1 wherein: the first animal identifier is unique within a set of all studies performed at a vivarium.

7. The marking code of claim 1 wherein: the combination of the first animal identifier and the second animal identifier is unique within a set of all animals in a vivarium and wherein neither the first animal identifier alone nor the second animal identifier alone is sufficient to uniquely identify one animal within the set of all animals in a vivarium.

8. The marking code of claim 1 wherein: the number of lines in the first data table is larger than a number of animal studies in a vivarium; and wherein at least one line in the first data table that is not at an index corresponding to a first animal identifier for a study in the vivarium comprises an error indicator.

9. The marking code of claim 1 wherein: the number of lines in the second data table is larger than the number of second animal identifiers in use in a first single animal study in a vivarium; and wherein at least one line in the second data table that is not at an index corresponding to a second animal identifier in use in the first single animal study comprises an error indicator.

10. The marking code of claim 1 wherein: the machine-readable code portion is free of any forward error correction.

11. The marking code of claim 1 wherein: the human-readable code portion is free of any characters dedicated to detection of valid versus invalid code reads of the human-readable code portion.

12. The marking code of claim 1 wherein: the human-readable code portion comprises an alphanumeric character set of no more than 32 characters, comprising 10 digits and 26 Roman alphabet letters minus at least four Roman alphabet letters.

13. The marking code of claim 1 wherein: the human-readable code portion comprises an alphanumeric character set wherein each character in the set is formed from a subset of seven connected linear line segments, wherein each line segment is either parallel to or orthogonal to all other line segments.

14. The marking code of claim 1 wherein: the machine-readable code portion comprises a visible spine axial to the tail of the rodent, and wherein each bar in the code portion is perpendicular to the spine and, unless a missing bar, is touching the spine; wherein the set of possible bars comprises (i) a descender bar, (ii) an ascender bar, (iii) a full height bar; (iv) a missing bar; and (v) an oversize bar; wherein: a descender bar is visible only on a first side of the spine; an ascender bar is visible only on a second side, opposite of the first side, of the spine; a full height bar is visible on both the first and second side of the spine; a missing bar is not visible or is negligibly visible; an oversize bar is longer than a full-height bar.

15. The marking code of claim 14 wherein: the oversize bar indicates a deleted bar position.

16. The marking code of claim 15 further comprising: an additional bar position for each deleted bar position.

17. The marking code of claim 16 further comprising: the additional bar position is at the end of the machine-readable code portion.

18. A method of marking rodents for identification using the marking code of claim 1.

19. A method of reading rodents identification using the marking code of claim 1.

20. The marking code of claim 1 wherein: the human-readable code portion comprises an alphanumeric character set wherein each character in the set is formed from a subset of seven connected linear line segments, plus an additional character comprising a fill region bounded by at least three linear line segments, wherein this character indicates a deleted character position, and wherein the human-readable code portion comprises an additional character position for each deleted character position.

21. The marking code of claim 1 wherein: the human-readable code portion comprises an alphanumeric character set wherein each character in the set is formed from a subset of seven connected linear line segments, plus an additional character comprising one or more diagonal line segments, wherein this character indicates a deleted character position, and wherein the human-readable code portion comprises an additional character position for each deleted character position.

22. A marking code for encoding information on a rodent's tail comprising: a human-readable code portion comprising a plurality of human-readable alphanumeric characters; a machine-readable code portion comprising a plurality of evenly spaced bars of varying length perpendicular to a tail of the rodent; wherein the human-readable code portion comprises a unique identifier for the rodent within the scope of a first study in a vivarium; wherein the machine-readable code portion comprises a unique identifier for the first study within the scope of the vivarium; and wherein the machine-readable code portion and the human-readable code portion are visibly marked on the rodent's tail.

23. The marking code of claim 22 wherein: wherein the human-readable code portion comprises a coding symbol set comprising a non-null subset of an alphanumeric symbol set comprising the digits 0 through 9 and the Roman alphabet; and wherein any symbols in the alphanumeric symbol set that are easily confused with one or more other symbols in the coding symbol set are not included in the coding symbol set.

24. The marking code of claim 23 wherein: the human-readable code portion comprises a delete symbol that is distinct from the superposition of all other symbols in the coding symbol set.

25. The marking code of claim 24 wherein: the human-readable code portion comprises a location reserved for a symbol whose location has been deleted by marking that location with the delete symbol.

26. The marking code of claim 22 wherein: the machine-readable code portion comprises a vine code.

27. The marking code of claim 26 wherein: the vine code comprises a marked continuous spine aligned with the tail axis.

28. The marking code of claim 26 wherein: the machine-readable code portion comprises a delete symbol that is distinct from the superposition of all other symbols in the vine code.

29. The marking code of claim 28 wherein: the machine-readable code portion comprises a location reserved for a symbol whose location has been deleted by marking that location with the delete symbol.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a rodent tail with a combined code tattooed.

(2) FIG. 2 shows prior art.

(3) FIG. 3 shows an exemplary vine code with a spine.

(4) FIG. 4 shows combined machine-readable and human-readable code combination.

(5) FIG. 5A shows an exemplary human-readable code.

(6) FIG. 5B shows an exemplary human-readable symbol set.

(7) FIG. 5C shows an exemplary set of capital letters missing from a human-readable symbol set.

(8) FIG. 6 shows a sparse index table.

(9) FIG. 7A shows a corrected error in a human-readable code.

(10) FIG. 7B shows an alternative corrected error in a human-readable code.

(11) FIG. 7C shows a corrected error in a machine-readable code.

(12) FIG. 8 shows an automated device for tattooing rodent tails.

DETAILED DESCRIPTION OF THE INVENTION

(13) Prior art barcodes use one code or marking to identify an object. Most often, the barcode is essentially a font or encodes a number, including a binary number. The characters encoded by the font or the number encoded is then the object identifier. Within some context or scope, that identifier is hopefully unique. Prior art barcodes generally assume reliable printing or marking, with some codes allowing for read errors by means of check digits or characters, parity, CRC, checksums, hash values, or forward error correction. Prior art also assumes a uniform or consistent surface on which the barcode is printed or placed.

(14) Animal ID, particularly for animals in a vivarium, including but not limited to rodents, have several unique requirements.

(15) First, marking may not be reliable. In fact, if marking is accomplished by tattooing then marking errors may occur more frequently than reading errors. Thus, it is desirable to have a code that allows for correction of mismarked code symbols.

(16) Second, not only is there limited space on which to place animal IDs, particularly on a rodent tail, but also the idealized codes space is large: being ideally globally unique codes for all animals for all time. These two conditions are in direct conflict, as a large code space requires a physically larger code as marked, for a given resolution and desired reliability.

(17) Third, accuracy of reading animal ID is critical in a vivarium. A single misread ID may invalidate a study or call into question the validity of a study. Thus, not only should reads be valid, but also error rates should be computable so as to provide a credible confidence level for both ID reads and study results.

(18) Fourth, there is typically a requirement for an animal code to be both human-readable and machine-readable. While such a requirement exists in many other barcode applications and is solved by simply replicated the data once as a machine-readable barcode and again as human-readable text, the space for marking animal ID is extremely limited and thus this space-inefficient solution is not practical.

(19) Fifth, the human-readable code should be one that is quickly and easily memorized by a human handler, who may well have both hands full and thus is not able to readily copy a read animal code onto paper or a keyboard. This requirement indicates a short code should be used, using symbology that is easy for people to memorize quickly and reliably. However, a short code is in direct conflict with desire to support a large code space.

(20) Sixth, a rodent tail is long and thin. Therefore, at least for the machine-readable code, a symbology that uses short-height symbols is desirable. While prior art barcodes, as parallel lines, support short symbols, prior art barcodes are inadequate as explained herein.

(21) Seventh, a common method of marking rodent tails is by use of tattoo. Tattoo heads mark points or lines at effectively a fixed width. Prior art barcodes typically use variable line width a part of the code. Such variable width lines are not compatible with standard tattoo heads.

(22) Eight, rodent tails are live and change length as the animal grows, and in addition, both grow unevenly and support significant live curvature that may rapidly vary in shape. Prior art barcodes typically use variable line spacing as part of the code. Such variable width line spacing is not compatible with realities of changing tail length, curvature and space due to growth and live animal activity.

(23) Ninth, controlling a tattooing head, with respect to a rodent tail, is difficult for both manual and automated operation. Therefore, the symbology used must mitigate these practical limitations. In particular, short line segments and diagonal lines are less reliably formed than perpendicular lines and line segments in one of a small number of fixed lengths.

(24) Tenth, because rodent tails changes shape rapidly, it is desirable to have a symbology that inherently provides local curvature information. Although humans are excellent at reading text on curves, automated systems are not. Prior art barcodes do not support dynamically curving substrates.

(25) Eleven, the width (diameter) of a rodent tale tapers from somewhat wide near the animal to narrow near the tip of the tale. Because of the very limited area in which to mark a rodent tale, it is desirable to have a code that takes advantage of the tapered aspect ratio of the rodent tail. Prior art barcodes assume a consistent, rectangular aspect ratio of the available substrate area.

(26) Twelfth, it is critical in vivarium applications that reading animal IDs is both highly reliable and that reliability (error rate) is computable. Some prior art uses check digits to catch errors. However, check digits are very inefficient. For example, a single check digit, assuming uniform and random errors, catches only 90% of errors, leaving a 10% uncaught error rate. To get a 0.1% (one out of one thousand) error rate it is necessary to use three check digits. If the animal code is three digits to start with, then the use of check digits doubles the necessary animal area. Some codes use error correction, such as parity or Reed-Solomon codes. Such error correction is most efficient for longer data blocks. The short data blocks of the animal ID do not support efficient error correction. For example, a 20-bit animal code would typically require another 14 bits for Reed-Solomon bits. Therefore, forward error correction is not an effective means to detect read errors.

(27) Thirteenth, many prior art methods of marking animals, including rodents, requires anesthetizing the animal prior to marking. Such anesthetizing is slow, expensive, stresses the animals, and requires specialized equipment. Placement of RFIDs or marking areas other than the tail typically requires anesthetizing. Some prior art marking methods, such as notching ears or punching in ear tags is painful for the animal and stresses the animal.

(28) Fourteenth, it is desirable to minimize the cost of marking each animal; minimize the need for specialized equipment; minimize reliance on a single vendor for equipment or supplies; minimize the need for specialized training of the humans doing the marking; and provide for high throughput of marking new animals. Prior art methods generally are time consuming to mark, making them costly; or use specialized equipment or vendor-lock-in equipment or supplies (including ink cartridges or pre-manufactured ear tags); or are difficult to perform, requiring specialize training, which also limits the number of personnel available for marking and also limits throughput. Specialized equipment may also limit throughput.

(29) Fifteenth, because it is desirable to keep code size at a minimum subject to all other requirements, it is desirable to be able to re-use animal codes, even if full animal IDs' are not reused. Prior art barcodes, being effectively fonts, do not permit effective index tables (such as in a database) that may be updated to permit re-use of animal codes. Note that we distinguish between the animal code, which is marked on the animal, and the animal ID, which is the ID associated with the animal for data recording and is unique within the scope of that animal ID. A table may be used to convert the animal code to the animal ID, where the index into the table is the animal code and the entry at that line of the table is the animal ID, as well as optional other data about either the animal code or the animal ID, or other related data. For this use, the animal code may also be called a token, and the table called a token table.

(30) Sixteenth, it is desirable to integrate the animal management, including marking and reading, with computer data about the animals, studies and vivarium. Prior art barcodes, being simple font representations of predetermined text strings or numbers, have no inherent integration. In addition, such integration is typically always required in a vivarium. Prior art codes do not take advantage of such integration to implement the desired features of animal making and animal codes, as described herein, or to mitigate weaknesses, as described herein. Therefore, it is desirable to take advantage of computer integration and computer databases as part of the animal marking and animal ID system.

(31) Seventeenth, prior art animal marking codes and methods often use small or extremely small code spaces. The minimum code space is marking uniquely each animal in one cage. For this purpose, sometimes colors are used. Five colors limit the number of animals in a cage to five. In addition, colors are typically not easily machine-readable. In particular, they cannot be read in infrared light, meaning they cannot be read during the normal nocturnal activity period of mice. Ear notches might have a code space of about 20, which is adequate for one cage. However, ear notches hurt the animals, and are also subject to a large human error rate as some notches may not be seen, or the ears may become damaged, or a human reader may mix up the two ears. Ear notches are very difficult to machine-read.

(32) Eighteenth, it is desirable to be able to have a human-readable code available to a human if both of the human's hands are full. This means that the need to hold a barcode reader to read a mouse ID is undesirable. In addition, the need for a barcode reader for every mouse interaction is time-consuming and requires a large amount of expensive, specialized equipment.

(33) Nineteenth, barcoded ear tags are undesirable because (i) they are expensive; (ii) they have single-vendor lock-in; (iii) they may not be available with the codes desired; (iv) they are not directly human-readable; (v) they require both operator hands to use (one to hold the mouse and one to hold the reader); (vi) they typically require the operation to be online to a computer in order to read the mouse ID. Paper and pencil may not be substituted, even temporarily; (vii) they may not be tolerated by the mouse; (viii) they may be chewed by other mice; (ix) relatively small amounts of detritus, such as food or blood, will make the code unreadable; (x) the animal will not do any self-cleaning of the ear tag. (However, they will self-clean their own tails.)

(34) Twentieth, and of particular importance, animal IDs in a vivarium are typically needed in two separate scopes: first within one study and second within the entire vivarium. A scope of a single cage is poor for several reasons. One is that the cage number must be used in order to make the mouse ID unique within a study. This introduces several sources of errors and also makes cage swapping, or moving mice between cages both cumbersome and error prone, as the indexing system must be properly and timely updated. It also means that mice from more than one cage cannot be intermixed. If they are accidentally intermixed, all those mice must be removed from the study immediately. Therefore, it is strongly desirable that a mouse ID be unique within a single study. Study sizes may vary from about five mice up to many hundreds. Therefore, a codes space of at least five hundred is typically necessary for animal ID within a study. Animals from different studies are almost never intermixed. Often the cages for one study are physically segregated from cages used in another study. Therefore, often knowing only the animal ID unique for one study is adequate for a large percent of animal handling and animal ID activities. It is advantageous to have the animal code unique to a study to be human-readable, as this meets many of the above numbered requirements or desires.

(35) With regards to the scope being the entire vivarium, it is necessary at various times and locations to assure than animals from one study are unquestionably part of that study, and that study animals are not intermixed, and that the absolute identity of every animal within a vivarium, or within a group of vivariums which might ever move or share animals, be known without risk of error, or with an extremely small and computable error. Thus, animals need to be permanently marked in some way with a code space equal to the maximum possible number of animals in a vivarium, and in addition provide some means of validating valid reads of this animal code. Such a code space is in the range of 100,000 to one million, and might even be larger in some applications.

(36) Novelty

(37) A novelty of embodiments is to use both a human-readable code portion and a machine-readable code portion. In one embodiment these two portions use different codes. In one embodiment, these animal code portions are marked on an animal's tail, such as a rodent tail. In one embodiment the animal code portions are tattooed. The tattooing may be manual or automated.

(38) FIG. 1 shows an exemplary human-readable code portion 11 and an exemplary machine-readable code portion 12 on a rodent tail 13. 10 is the body end of the tail; 13 is the tip end of the tail. 14 shows a partial rodent body

(39) In some embodiments the human-readable code portion is also machine-readable.

(40) In some embodiments, the human-readable code portion is unique within one current study, however, such a limitation is not a requirement for all embodiments.

(41) In some embodiments, the machine-readable code portion is unique for all animals in a vivarium, or for all currently live animals in a vivarium. In some embodiments, the combination of the human-readable code portion and the machine-readable code portion is unique for all animals in a vivarium, or for all currently live animals in a vivarium. For example, the machine-readable code portion may map to a study ID.

(42) In some embodiments an animal code, the data marked on the animal, is mapped via a mapping table to the animal ID. Such a mapping table is shown in FIG. 6. This may be true for just the human-readable portion, or just the machine-readable portion, or both. One advantage of this is that animal codes may be reused, while keeping the animal IDs unique, at least within a known scope, context, and time period. For such embodiments, the animal code may be called or thought of as a token. The animal code, or token, is then used as (or is) an index into one or more mapping tables. One such map type is a through map meaning that some or all of the table lines comprise data that is the same as the index for that line. In these cases the code is then the same as the corresponding ID.

(43) Continuing with FIG. 1, we see both a human-readable code portion 11 and a machine-readable code portion 12 marked on a rodent tail. 13 The rodent may be a mouse. The method of marking may be tattooing. The human-readable code 11 may comprise an alphanumeric character set comprising digits and Roman letters. The human-readable code may comprise digits, as shown 11 in this Figure. The human-readable code 11 may comprise an alphanumeric character set comprising digits and limited Roman letters such that the remaining Roman letters in the character set are easily recognizable and clearly distinguishable from digits or other Roman characters. The human-readable code portion may be one character, two characters, three characters, four characters, or five characters, or the range of 1-2, 1-3, 1-4, 2-3, 2-4, or 3-4 characters.

(44) Some embodiments use symbol sets that are numeric only or alpha only. Symbol sets may comprise or consist of foreign (non English) alphabet symbols. Such symbol sets are expressly claimed. Numbers instead of alphanumeric human readable symbol sets do not code as efficiently with respect to animal tail area, and thus are not preferred. Also, they may be more difficult of humans to read and memorize accurately. (Compare, for example, quickly memorizing five digits compared to three letters.)

(45) The human-readable code portion 11 may code an animal ID, unique within a known scope. The human-readable code portion 11 may code an animal ID, unique within one animal study. The human-readable code portion 11 may be a first animal code, where the first animal code must be looked up in a first animal code-mapping table whose entries comprise the corresponding animal ID.

(46) Continuing with FIG. 1, the machine-readable code portion 12 may be a vine code, with or without a spine. It may be a barcode known in the art. The spine is aligned with the tail such that the code is easy to read with machine vision, as the spine clearly demarks the location and curvature of the tail, and also provides a reference point for determining if the bars, (or leaves) in the code are ascenders, descenders, full-height or missing. A rodent tail may not only curve dynamically with rodent activity, but also rotate around the tail axis. The tail may also curve upward. Thus, a code spine is valuable to determine all such distortions from an ideal, flat, rectangular, fixed substrate for the code marking. In addition, the tail may be partially covered with detritus, such as bedding or other material. A continuous spine makes if far easier to determine reliably if a portion of the machine-readable code is so obscured.

(47) Prior art required white or black cage bottoms to machine-read a tail barcode. Prior art required cages to be free for bedding, as that might obscure a portion of the code, causing a misread. Use of a vine code with a spine eliminates these requirements as the dark markings on a light tail provide sufficient contrast, and use of a spine does not require that the edges of the tail be identifiable.

(48) In some embodiments the machine-readable code portion 12 is an animal identifier within a known scope. In some embodiments the scope is all the animals in a vivarium. In some embodiments the scope is all the live animals in a vivarium. In some embodiments the scope is all the animals in a group of vivariums.

(49) The machine-readable code portion 12 may be a second animal code, where the second animal code must be looked up in a second animal code-mapping table whose entries comprise the corresponding animal ID. The machine-readable code may be a second animal code, where the second animal code must be looked up in a second animal code-mapping table whose entries comprise a corresponding study ID.

(50) In some embodiments the combination of the human-readable and the machine-readable code portions is an animal identifier within a known scope. In some embodiments this scope is all the animals in a vivarium. In some embodiments the scope is all the live animals in a vivarium. In some embodiments the scope is all the animals in a group of vivariums. In some embodiments the scope is all the live animals of all vivariums. In some embodiments the scope is all the animals of all vivariums.

(51) The combination of the human-readable and the machine-readable code portions may be a third animal code, where the third animal code must be looked up in a third animal code mapping table whose entries comprise the corresponding animal ID.

(52) The code space of the machine-readable code portion may be no more than 10,000; 100,000; 1,000,000; 10,000,000; or 100,000,000. The code space of the machine-readable code portion may be no more than 12 bits, 14 bits, 16 bits, 18 bits, 20 bits, 22 bits 24 bits, 26 bits, or 28 bits. The code space of the machine-readable code portion may in the range of 12-14 bits, 12-16 bits, 14-18 bits, 16-20 bits, 18-22 bits, 20-24 bits, or 22-28 bits, 14-20 bits, or 16-28 bits. The machine-readable code portion may or may not comprise check bits, parity bits, checksums, CRC, and forward error correction.

(53) Turning now to FIG. 2 we exemplary prior art. There is no visible spine. An ascender bar or leaf is shown 10. A descender leaf is shown 21. A full-height leaf is shown 22. A minimal leaf is shown 23. The machine-readable code in this Figure is made human-readable by replicating the data in the code as text, 24.

(54) FIG. 2 shows prior art vine code. This is an example of one of a number of postal code symbologies. Prior art vine code symbologies did not have a marked spine. Prior art vine code symbologies have no provision for marking a bar as deleted, and have no provision for adding a replacement bar for a deleted bar. Prior art vine code symbologies use pre-determined encodings and do not use a dynamic lookup table to map from a printed code to an effective identifier of the object (or substrate) on which the code is printed.

(55) Turning now to FIG. 3, we see an exemplary vine code of one embodiment. 30 is a full-height leaf. 31 is an ascender leaf. 32 is a descender leaf. Location 33 shows a missing leaf. 34 shows a fixed-width gap between leaves. Missing leaves, such as 33, are detected by noticing that the gap between adjacent leaves is twice the fixed-width gap 34 plus the width of one bar. Note that the prior art code in FIG. 2 does not have missing leaves, but rather minimum height leaves.

(56) In some embodiments, the leaves (or bars) of the vine code may encode two bits each. In some embodiments, two bits may be encoded as one of: (i) an ascending leaf 31; (ii) a descending leaf 32; (iii) a full-height leaf 30; (iv) a missing leaf 33. In some embodiments a leaf that is longer than a full-height leaf may be used to indicate a deleted leaf position. In this Figure, line width is not part of the coding. In this Figure, the spacing between leaves 34 is fixed: line spacing is not part of the coding. An advantage over prior art is that by ignoring both line width and line spacing in the coding, the significant real-time and growth-based dynamics of a rodent tail do not distort the marked code so as to reduce the reliability of reads or cause misreads. Even if any such misread rate is low, it is not accurately computable. In addition, both line width and line spacing are difficult to control when tattooing, whether the tattooing is done manually or by a robot.

(57) An exemplary vine code, such as shown in FIG. 3, may encode 12-14 bits, 12-16 bits, 14-18 bits, 16-20 bits, 18-22 bits, 20-24 bits, or 22-28 bits, 14-20 bits, or 16-28 bits. The vine code may or may not comprise check bits, parity bits, checksums, CRC, and forward error correction.

(58) Turing now to FIG. 4, we see an exemplary combination of both a human-readable code portion 40 and a machine-readable code portion 41.

(59) Turning now to FIGS. 5A and 5B, we see an exemplary symbology for the characters in a human-readable code. All characters may be formed by a non-empty subset of seven straight-line segments, arranged in a well-known seven-segment formation. The use of this formation is uniquely advantageous for tattooing rodent tails because each line segment may be formed by motion in only one axis. The motion may be manual or automated. Either the rodent (or the platform on which the rodent rests) may be moved, or the tattooing apparatus may be moved. In one embodiment, motion along the axis of the tail is implemented by supporting a tattooing head on a linear track, which in turn may be supported by a gantry or other support structure. In one embodiment, the other axis may be implemented by rotating the tail around the tail axis. For manual tattooing, such rotation may be via one or more finger(s) gently placed on the tail. For automatic tattooing, a motorized rubber finger or fingers may be used, effectively operating similar to manual operation. Tattooing movement in only one axis at a time is far easier for human operators, which is a unique problem, and for which this code symbology is a novel solution.

(60) FIG. 5A shows an exemplary code, here the three characters 348, using the exemplary symbol set shown in FIG. 5B. Note that all symbols shown in FIG. 5B may be formed from a subset of the seven-segment line segments. Note that all symbols shown in FIG. 5B are readily identifiable without confusion with other symbols, and also readily memorized in codes.

(61) The symbol set for the human-readable code may comprise an alphanumeric character set comprising digits and Roman letters such as shown in FIG. 5B. The human-readable code may comprise digits, as shown in FIG. 5A. The human-readable code may comprise an alphanumeric character set comprising digits and limited Roman letters such that the remaining Roman letters in the character set are easily recognizable and clearly distinguishable from digits or other Roman characters. The human-readable code portion may be one character, two characters, three characters, four characters, or five characters, or the range of 1-2, 1-3, 1-4, 2-3, 2-4, or 3-4 characters. One such reduced character set is shown in FIG. 5B. Here, the letters D and N are represented more closely to their traditional lower case appearance: d and n respectively. The letters T and Y appear asymmetric in the code; however, they are still easily identifiable by people. Within the context of our claims, we explicitly claim all similar reduced character sets, similarly reduced for readability in a seven-segment format, under the Rule of Equivalents.

(62) FIG. 5C shows Roman characters not used in the exemplary symbol set due to their similarity to other digits or characters.

(63) Another embodiment uses the seven line-segment basis expanded to nine segments by the addition of two, three, or four diagonal segments in the upper and lower rectangles formed within the digit 8. The addition of these diagonal segments permits the full 36-character alphanumeric character set to be distinctly marked.

(64) Numeric only and alpha only symbol sets are expressly claimed as alternative embodiments to an alphanumeric symbols set.

(65) Turning now to FIG. 6, we see an exemplary embodiment of a mapping table. Such a table may be used only for mapping the human-readable code portion, or mapping only the machine-readable code portion, or two separate tables with one for each code portion, or a single unified code table that maps both code portions with one lookup in that table. The table is shown 71. An index, such as an animal code, is shown 70. Here the animal code is 348. This index points to table line 72. Here we see that this table line 72 comprises an animal ID of 4021. Table lines that do not have animal codes pointing to them comprise an error indicator, as shown for two table lines, 73.

(66) A particular advantage, in some embodiments, of using one or more mapping tables, is that only currently valid animal ID are entered into the table. Any animal code properly read will then map to a valid animal ID. However, a misread of an animal code then has (an accurately computable) chance of mapping to a table line that does not contain a valid animal ID, but rather contains indication of error, that is, a misread. We refer to such a table entry as an error code. For example, if one study uses 15 animals, and the code space for the animal ID for animals within one study is 625, then random misreads will be caught (62515)/625=97.6% of the time. This method of catching read errors does not require any additional marking space on the animal. It is desirable to assign the animal codes, or tokens, using some algorithm, such as pseudo-random assignment (or one of many other known algorithms, such as incrementing token numbers by a relatively prime), to cause misreads to be effectively random, which assures that such reliability computations are applicable. Ideally, both the codes and IDs are distributed throughout their corresponding code and ID spaces.

(67) This method of detecting errors may also be used, in some embodiments, in the machine-readable code portion. For example, if the machine-readable code portion is 28 bits, providing a code space over 250 million, and the total vivarium population is 100,000, then random misreads will be caught (268,435,456100,000)/(268,435,456)=99.96% of the time. In practice the error rate is likely better than this, because both the machine-readable portion and the human-readable portions must both map to valid table entries. Equivalently, if either the human-readable portion or the machine-readable portion generates an error, then both the human-readable portion and the machine-readable are questionable, even though one of them may in fact have been read correctly.

(68) The table in FIG. 6 also demonstrates a study ID table using machine-readable codes as the index. For this interpretation of FIG. 6, the machine-readable code is shown as 70, and the study ID is 4021 in line 72.

(69) There are three key purposes and benefits of such tables. First, by using a sparse table with respect to the number of valid animal IDs, a large fraction of read errors of the animal code will be caught because they map to table lines that contain an error code, rather than a valid animal ID. One such claim is explicitly such a sparse table and also one such claim is the use of the sparse table for misread detection. Second, the mapping table provides the ability to reuse animal IDs, treating them in one interpretation as re-usable tokens. When an animal is no longer alive, or other wise cannot have its animal code read, that animal code may immediately be reassigned to another animal. Such reassignment is implemented by changing the data in the table line indexed by that animal code. This approach allows a smaller code space for the animal codes on the animals, while permitted a large code space for the animal IDs. It simultaneously permits most misreads to be caught without the space-consuming need for check characters on the animal. In addition, shorter character sequences are memorized by humans more reliably than longer character sequences, and are thus more reliable. The third benefit of the mapping tables is the inherent integration or ability to integrate with other animal, study, and vivarium data. Such data may be additional information in the table lines or otherwise in accessible databases indexed by either the animal code or the animal ID. Such integration provides faster and more reliable data entry and will typically catch errors so quickly as to permit immediate correction without damage to the study or increased costs, or less reliable study data. We specifically claim the use of the mapping tables to implement the processes, procedures, applications, and benefits as described herein.

(70) Turning now to FIG. 7A we see one embodiment of correcting a marking error in a human-readable code. The correct code is 348. However, here the 4 in location 101 was marked incorrectly. The correction is made by placing a large X, or another marking that is distinct from, or larger than, or both, the superposition of all other characters in the symbol set. Such a deletion indicator is shown as 101. Then, the correct deleted symbol is added at the end of the code, shown here at location 100. Note that this method may be used for more than one error symbol in a code. Locations other than at the end of the code may be used. However, at the end is preferable because the error may not be discovered until after the code is otherwise completed.

(71) In FIG. 7B we see an alternative embodiment and method of correction. Here location 102 shows the location of the error. The deletion mark is a large black area. Here the deletion mark comprises markings not found in any symbol. The correct symbol is placed at the end of the code, shown here at location 103. The correct code is 348.

(72) In FIG. 7C we see one embodiment of correcting a marking error in a machine-readable code. Location 104 shows an extra long leaf, or bar. This extra-long leaf will cover any leaf that is an ascender or descender. The correct leaf is then placed at the end of the code, here shown at location 105. More than one leaf in error may be corrected this way. For a leaf or leaves that are not at the correct spacing, such leaves are covered with the extra-long leaf, and also an extra-long leaf is placed at the nominal correct location or locations. Then, as before, the deleted leaves are placed at the end of the code.

(73) For a machine-readable vine code as shown in FIG. 7C, an alternative embodiment for some marking errors is to assign the machine-readable code of the as marked code, to the same ID, such as a study ID, as the correct machine-readable code, by updating the machine-readable mapping table.

(74) In FIG. 8, we see an automatic machine suitable for marking the codes of this invention, or many other codes. It is particularly suitable for the codes described herein. 91 is a mouse, as an example of animals or rodents. The mouse may be restrained by a mouse-restraint, not shown. The mouse tail is 98. The mouse rests on a mouse table, 92. The mouse table 92 has an optional valley, groove or dip, 95. 93 shows the tip of a tattooing head, or tattoo gun, 96. The tattooing head is attached via a traveler 97 to the beam of a gantry 90. One or more movable fingers 99 hold the mouse tail 98 in position.

(75) To mark the human-readable and machine readable codes described as embodiments herein, two directions of motion of the tattoo tip 93 relative to the mouse tail 98 are required. The first direction of motion is along the axis of the mouse tail, shown by arrow 201. This motion makes horizontal lines segments using code orientations shown in the other Figures. The second direction of motion is the rotation of the mouse tail 98 around its axis. This rotation is accomplished by a simple rolling motion of the tail, performed by moving the fingers 94 in the direction shown by the arrow 99. There may be one or more fingers; two fingers are shown here. An alternative to finger(s) is one more capstans in contact with the tail. Yet another alternative is to move the mouse or the mouse table. This alternative may use a linear motion. The valley or groove 95 keeps the mouse tail straight, although in some embodiments this valley is not required. A tail clamp may be used. A key advantage of the symbologies of embodiments of this invention is that only one direction of motion is required at a time. This is ideal for both human tattoo operators and for the machine of this Figure.

(76) The machine requires one additional motion. This is the raising and lowering of the tattoo tip, 93, relative to the mouse tail, 98. This motion is shown by arrow 200. This may be implemented using a solenoid, motor, or hydraulics. Either the tip 93 alone, or the tattoo head 96, or beam of the gantry 90 may be moved. Rotation may be used, rather than linear motion, because the distance the tattoo tip has to travel is small.

(77) One additional motion may be automatic by the machine or may be manual. That is the raising and lower of the fingers 94 in order to place and remove the mouse. The fingers may be spring-loaded, have a two-position latch, slide in and out of finger holders. Many other implementations are possible, as those in the art know.

(78) The machine may have additional motions. The mechanical design in this Figure is highly schematic, as those in the art appreciate. Fingers may move along the mouse tail, or more than one finger or set of fingers may be used.

(79) Embodiments may move the mouse table, 92 in one or more motions in place of the motions described.

(80) It is desirable in some embodiment, to be able to slide the traveler 97 along the gantry beam by hand in order to position the tattoo head 96 relative to the tail 98.

(81) Motion 201 may be accomplished via stepper or servo motors and a belt, cable or chain, or by the use of driven screw, or a capstan or by hydraulics, or by other means, known in the art. Motion 99 may be accomplished by a solenoid, motor, or hydraulics, or other means, known in the art.

(82) Both control of the operation of the machine of FIG. 8 and the verification of correctly marked codes may be performed by the use of a single video camera, not shown in this Figure. Such use of a single camera for both purposes is claimed as an embodiment.

(83) An alternative embodiment to traditionally tattooing is to damage, abrade or puncture the skin on the rodent tail either before after applying ink. Such action may be caused by an automatically controlee needle or knife, or a laser. In one embodiment the ink is applied manually, in bulk, either before or after damaging, abrading or puncturing the skin in the pattern to be marked on the tail.

(84) An alternative embodiment to traditionally tattooing is using an ink jet printer print head to apply ink. The skin of the rodent tail may be damaged, abraded or punctured prior to applying the ink via the print head. One advantage of using an ink jet print head is the ability to easily use colored ink. Such an embodiment is specifically claimed. In one embodiment this color is used to identify a mouse within the scope of one cage. The color may be part of one of the code portions described herein, or may be an additional marking.

(85) The human-readable code and its associated symbology, features and data table are claimed as embodiments independent of the machine-readable code.

(86) The machine-readable code and its associated symbology, features and data table are claimed as embodiments independent of the human-readable code.

(87) An embodiment comprises reading codes as described herein, including devices of reading and methods of reading. Embodiments include systems that use both reading and writing, which might be use, for example, as a comprehensive animal identification system in a vivarium.

(88) A reading embodiment includes one that tracks the drift of the appearance of a marked code over time. Such changes include: stretching either lengthwise or width as the animal tail grows; distortion, such as adding curvature as the animal tail grows, ages, or distorts; change in color or contrast of the marking or the tail skin over time; blurring of lines over time; missing sections of line segments due to injury, shedding, cleaning and the like. On one embodiment such changes are tracked via one or metrics built into the native reading of a code. One such metric is the non-uniform stretch of bar spacing. For example, a tail does not grow uniformly. Thus, one end of a bar code may stretch more than the other end. Therefore, it is desirable to track such non-uniform stretching as the rodent grows and use that tracked metric (which may be averaged, or predicted) as a baseline of expected positions of symbol elements. Another example is the aspect ratio of either alphanumeric symbols of vine code symbols changing over time. Therefore, it is desirable to track such aspect ratio changes as the rodent grows and use that tracked metric (which may be averaged, or predicted) as a baseline of expected shapes of symbol elements. Such tracking and use of time-evolving metrics, tracked per animal, is expressly claimed.

(89) Another aspect of change to animal markings consists of less-structured changes. For example, an injury or detritus may add a dark or light area to section of a marked code or the background tail skin. A recorded image of the rodent tail, including the section with the marked code, may then be used as a baseline for code recognition. For example, a permanent piece of detritus, or an injury, might obscure a section of code. Although that code section is no longer readable using the native code reading algorithm, by use of a recorded image of that animal tail, plus reading the remaining portions of the code, the animal can be uniquely and reliably identified. Thus, in this aspect, images of animal markings are recorded and tracked as they change over time, and these images (which may averaged) are used as part of animal code reading and animal ID determination. This aspect and these reading steps are expressly claimed as embodiments for both devices and methods.

Additional Embodiments

(90) Embodiments of this invention explicitly include all combinations and sub-combinations of all features, elements and limitation of all claims. Embodiments of this invention explicitly include all combinations and sub-combinations of all features, elements, examples, embodiments, tables, values, ranges, and drawings in the specification and drawings. Embodiments of this invention explicitly include devices and systems to implement any combination of all methods described in the claims, specification and drawings.

(91) Embodiments specifically include all independent claim limitations in all combinations with all independent claims.

(92) Embodiments are claimed wherein the word comprises is replaced with the word, consists, in one or more places in a claim.

Definitions

(93) Communicationmay be electromagnetic, optical or audio. Audio comprises sub-audio and ultrasonic audio.

(94) Gantryany support for a linear track, including the track itself, or an open support or closed support. May include a rectangular enclosure or a system adapted to be secured in an existing support structure (such as a bench or fume hood).

(95) IR LEDAn IR LED is one in which its spectrum peaks or is predominantly in the infrared spectrum, as normally defined with respect to IR LEDs and IR cameras. Note there may be some overlap between the IR spectrum and the visible light spectrum.

(96) Pathogen-freemeans the population of microbes, including but not limited to bacteria, viruses, prions and toxins, relevant to the experiment, are sufficiently reduced to meet the needs of the study, or to not impact the health, performance or behavior of the target animal population or of the workers.

(97) Primary cage or home cagethe cage in which an animal spends more time than any other cage. Of note, there is a related term of art: home cage. The definition of primary cage is, in some embodiments, the home cage. An aspect of home cage/primary cage deals with the fungibility of the actual cage itself. Each time a cage is changed, the physical cage is generally either disposed of or removed for washing, and replaced by a clean cage. The new physical cage is considered the same primary cage. A primary cage may sometimes be distinguished from a non-primary cage by the purpose of the cage. For example, a home cage may be for living in, as compared to an experimental cage to which the animal is transferred that is equipped or located for one or more particular experiments for the applicable study.

(98) The primary cage is different from special purpose, behavioral-measurement, behavioral-detection, or behavioral-observation cages that are generally used for only a short time for the duration of a particular test due to cost and mindset.

(99) Sealed enclosurean enclosure sealed against pathogens that impact or alter study results, or alter the credibility or repeatability of study results, entering or leaving the enclosure.

(100) Sensormay or may not include the use of local or remote processors, and may or may not include local or remote software executing on the local or remote processors. Sensors may or may not communicate over a network. Multiple sensors may or may not include common elements.

(101) Setone or more, unless stated otherwise.

(102) Sparse tablean electronic table where not every table line contains an entry for the primary data either as an index or as table line data. However, otherwise empty or unused table lines may contain an indication of error.

(103) Subsetone or more elements of a set, including all elements of the set, unless stated otherwise.

(104) Sterilepathogen-free.

(105) Visible lightFree of visible light means the ambient light is sufficiently low and in a spectrum such that the animal's physiological state and behavior are consistent with its natural physiological state and behavior at night.

(106) White lightLight with wavelengths across the visible spectrum that is suitable for seeing color using a conventional color camera, as conventionally defined in the art. Note that white light may include some wavelengths outside of the visible spectrum.

(107) Ideal, Ideally, Optimum and PreferredUse of the words, ideal, ideally, optimum, optimum, should and preferred, when used in the context of describing this invention, refer specifically a best mode for one or more embodiments for one or more applications of this invention. Such best modes are non-limiting, and may not be the best mode for all embodiments, applications, or implementation technologies, as one trained in the art will appreciate.

(108) May, Could, Option, Mode, Alternative and FeatureUse of the words, may, could, can, option, optional, mode, alternative, typical, ideal, and feature, when used in the context of describing this invention, refer specifically to various embodiments of this invention. Described benefits refer only to those embodiments that provide that benefit. All descriptions herein are non-limiting, as one trained in the art appreciates.