Medical diagnosis assistance method

Abstract

Disclosed is a method for assistance with the establishment of a diagnosis of a patient, starting from at least one identified sign, and based on a computerized knowledge database including a medical ontology. The medical ontology includes: a list of signs forming a “sign” class; a list of pathological states forming a “pathological state” class; and a first set of logical relationships between the signs and the pathological states, each logical relationship establishing a correlative link between a sign and a pathological state. The method includes: a step of searching for potential pathological states, linked to at least one of the identified signs by the first set of logical relationships; and a step of identifying potential signs in which, for each potential pathological state, all of the signs linked by a correlative link to the potential pathological state are identified by the first set of logical relationships.

Claims

1. A device-implemented method for assistance with the establishment of a diagnosis of a patient, comprising: providing a device that incorporates a computerized knowledge database (10) of a medical ontology (12) that includes a first list (121) of signs (Si) forming a “sign” class, a list (122) of pathological states (Dj) forming a “pathological state” class, a first set (124) of logical relationships between the signs and the pathological states, each logical relationship of said first set (124) establishing a correlative link between a sign and a pathological state, a second list (123) of technical methods for acquiring medical images (Mk) forming an “imaging” class, each of said technical methods for acquiring medical images (Mk) defining a type of technology for acquisition of medical images and a positioning of the medical image on the patient, and a third set (126) of logical relationships between the signs and the technical methods for acquiring medical images (Mk), each logical relationship of said third set (126) defining a technical method for acquiring a medical image that makes possible an observation of a sign; receiving, at the device from an operator of the device, one or more identified signs; searching (503), carried out by the device, for potential pathological states, in which all of the pathological states linked by a correlative link to at least one of the identified signs are sought in the pathological state class by means of the first set (124) of logical relationships, said pathological states forming the potential pathological states; identifying (504), carried out by the device, potential signs in which, for each potential pathological state, all of the signs linked by a correlative link to said potential pathological state are identified in the sign class by means of the first set (124) of logical relationships, said signs forming the potential signs; and identifying (505), carried out by the device, technical methods for acquisition of interest, in which for each potential sign, all of the technical methods for acquiring medical images that make possible to observe said potential sign are identified in the imaging class by means of the third set (126) of logical relationships, said technical methods for acquiring medical images forming the technical methods for acquisition of interest.

2. The method according to claim 1, wherein the first set (124) of logical relationships comprises logical relationships according to each of which a sign (Si) is indicative of a pathological state (Dj).

3. The method according to claim 2, wherein the medical ontology (12) also includes a second set (125) of logical relationships between pathological states of the pathological state class, each logical relationship of said second set (125) defining a kinship link between a given pathological state and a parent pathological state of the given pathological state, wherein the step (503) of searching for potential pathological states includes the sub-steps of: searching for potential pathological states of order one, in which all of the pathological states indicated by at least one of the identified signs are sought, said pathological states forming the potential pathological states of order one, and searching for potential pathological states of order two to N, where N is a natural integer greater than or equal to two, the search for potential pathological states of order n, where n is a natural integer comprised between two and N, comprising, for each potential pathological state of order n−1, searching for all of the parent pathological states of this potential pathological state of order n−1 by means of the second set (125) of logical relationships, said parent pathological states forming the potential pathological states of order n, and wherein the step (504) of identifying potential signs includes identifying all of the signs indicative of a potential pathological state of order one and/or a potential pathological state of order n.

4. The method according to claim 3, wherein the step (504) of identifying potential signs includes classifying said potential signs, according to which all of the potential signs indicative of a potential pathological state of order m, where m is a natural integer comprised between one and N, form the potential signs of order m.

5. The method according to claim 1, wherein the step (505) of identifying technical methods for acquisition of interest includes classifying the technical methods for acquisition of interest, according to which all of the technical methods for acquisition of medical images making it possible to observe a potential sign of order m form the technical methods for acquisition of interest of order m.

6. The method according to claim 5, further comprising: displaying the technical methods for acquisition of interest.

7. The method according to claim 5, further comprising: constructing (508) an examination protocol, in which the technical methods for acquisition of interest are hierarchized according to their order, the technical methods for acquisition of interest of order m having priority over the technical methods for acquisition of interest of order m+1.

8. The method according to claim 7, further comprising: displaying the technical methods for acquisition of interest.

9. The method according to claims claim 7, wherein the technical methods for acquisition of interest of a same order are hierarchized according to their occurrence among all of the technical methods for acquisition of interest of the order, or among all of the technical methods for acquisition of interest of orders different from the order.

10. The method according to claim 1, further also comprising: displaying the technical methods for acquisition of interest.

11. The method according to claim 1, further comprising: selecting (510) technical methods for acquisition to be implemented, in which at least one of the technical methods for acquisition of interest is selected, all of the technical methods for acquisition of interest selected forming the technical methods for acquisition to be implemented, and one or more medical images, each medical image being acquired according to one of the technical methods for acquisition to be implemented.

12. The method according to claim 1, wherein the computerized knowledge database (10) also includes a library (11) of medical images, each medical image representing a sign, and wherein the method further comprises, after the step (504) of identifying potential signs: searching (511) for reference images in which, in a case of at least one of the potential signs, all of the medical images representing said at least one potential sign are sought in the library (11) of medical images, said medical images forming the reference images.

13. The method according to claim 12, wherein the step (511) of searching for reference images is carried out after the step (505) of identifying technical methods for acquisition of interest, and comprises, in the case of at least one of the potential signs, searching for medical images representing said at least one potential sign and being acquired by means of one of the technical methods for acquisition of interest.

14. The method according to claim 1, wherein the medical ontology (12) relates to ectopic pregnancies.

15. A non-transitory computer-readable medium on which are stored computer-executable instructions that, when executed by a computer, cause the computer to perform the method according to claim 1.

Description

DESCRIPTION OF THE FIGURES

(1) Other advantages and features of the invention will become apparent on reading the detailed description of implementations and embodiments that are in no way limitative, with respect to attached drawings, in which:

(2) FIG. 1 diagrammatically represents an example of a computerized knowledge database on which the method for assistance with the establishment of a diagnosis according to the invention is based;

(3) FIG. 2 represents an example of medical ontology;

(4) FIG. 3 shows an example of logical relationships between a pathological state and signs;

(5) FIG. 4 shows an example of logical relationships between several pathological states;

(6) FIG. 5 represents an example of a method for assistance with the establishment of a diagnosis of a patient according to the invention.

DESCRIPTION OF EMBODIMENTS

(7) As the embodiments described below are in no way limitative, it is possible, in particular, to consider variants of the invention comprising only a selection of characteristics described hereinafter, in isolation from the other characteristics described (even if this selection is isolated within a phrase containing these other characteristics), if this selection of characteristics is sufficient to confer a technical advantage or to differentiate the invention from the state of the prior art. This selection comprises at least one preferably functional, characteristic without structural details, or with only a part of the structural details if this part alone is sufficient to confer a technical advantage or to differentiate the invention from the state of the art.

(8) It is recalled that the term “sign” denotes any phenomenon linked to the state of a patient and covers, in particular, the concepts of “functional sign”, “clinical sign” and “paraclinical sign”. The term “pathological state” relates to a state of deterioration in the functions, in the morphology or in the health of an organ or organism, being manifested by the presence or absence of one or more signs. The term “technical method for acquisition of a medical image”, denotes a set of technical features making it possible to obtain a medical image capable of being compared with a reference image or with reference data.

(9) FIG. 1 diagrammatically represents a computerized knowledge database on which the method for assistance with the establishment of a diagnosis according to the invention is based. The knowledge database 10 is for example stored in a memory of a computer or server, or on a removable storage medium, such as a compact disc (CD) or a USB stick. It comprises a library 11 of medical images and a medical ontology 12. Each medical image represents one or more signs, and is associated with a label indicating different properties of the medical image. The label indicates, for example, the technique for acquisition of the medical image (CT or computed tomography imaging, ultrasound, etc.), the organ of the patient represented, the positioning of the image relative to the patient, for example the cross-sectional plane in which the image is situated, in the case of a plane image, or the borders of the image, in to the case of a three-dimensional image. The label can also indicate the presence of each sign that can be observed in the associated image, and the clinical state of this sign. The clinical sign can, for example, take one of the following values: “normal state”, “state varying from the normal”, “malformation”, “inflammation”, “excrescence” etc. The library 11 of medical images comprises, for example, pairs or sets of medical images, representing a given sign for different clinical states. In particular, one of the images can represent the sign for a state considered as normal.

(10) FIG. 2 represents the medical ontology 12 in greater detail. The medical ontology 12 comprises a list 121 of signs, a list 122 of pathological states, and a list 123 of technical methods for acquisition of medical images. For the sake of ease of reading, the signs are denoted individually or globally by the reference S.sub.i, where the subscript i is a natural integer comprised between one and P, the pathological states are denoted individually or globally by the reference D.sub.j, where the subscript j is a natural integer comprised between one and Q, and the technical methods for acquisition of medical images are denoted individually or globally by the reference M.sub.k, where the subscript k is a natural integer comprised between one and R. The integers P, Q and R are natural integers greater than or equal to two. They can be of the order of a few tens, a few hundred or even several thousand. All of the signs S.sub.i form a “sign” class, all of the pathological states D.sub.j form a “pathological state” class, and all of the technical methods for acquisition of medical images form an “imaging” class.

(11) In addition, the medical ontology 12 comprises a first set 124 of logical relationships between the signs S.sub.i and the pathological states D.sub.j, a second set 125 of logical relationships between the pathological states D.sub.j, and a third set 126 of logical relationships between the signs S.sub.i and the technical methods for acquisition of medical images M.sub.k. Each logical relationship of the first set 124 establishes a correlative link between a sign S.sub.i and a pathological state D.sub.j, as explained in more detail with reference to FIG. 3. Each logical relationship of the second set 125 defines a kinship link between a given pathological state D.sub.j, and a parent pathological state D.sub.l of this given pathological state as explained in more detail with reference to FIG. 4. Each logical relationship of the third set 126 defines a technical method for acquisition of a medical image M.sub.k making it possible to observe one of the signs S.sub.i. A sign S.sub.i can be observable by means of several technical methods for acquisition of a medical image M.sub.k. Moreover, a technical method for acquisition of a medical image M.sub.k can make it possible to observe several signs S.sub.i.

(12) FIG. 3 shows an example of logical relationships between a pathological state D.sub.1 and signs S.sub.1, S.sub.2, and S.sub.3, in the field of gynaecology. By way of example, it is considered that the pathological state D.sub.1 is an ectopic pregnancy and that the signs S.sub.1, S.sub.2 and S.sub.3 are respectively an empty uterus, a latero-uterine mass, and peritoneal effusion. Each sign S.sub.i is indicative of the pathological state D.sub.1. The logical relationships 31, 32, 33 are thus defined as a consequence. It is possible to assign to each logical relationship a coefficient representative of the probability that a sign is associated with a pathological state. During the execution of the method according to the invention, potential pathological states can thus be prioritized over others.

(13) FIG. 4 shows an example of logical relationships between several pathological states in the field of foetal medicine. By way of example, it is assumed that the pathological state D.sub.1 is CHARGE syndrome, and that it is a pathological state of order one. This pathological state D.sub.1 is the daughter pathological state of three pathological states D.sub.2, D.sub.3 et D.sub.4, corresponding respectively to a syndromic anomaly of the development of the eye, a syndromic genetic deafness, and an otorhinolaryngological malformation. These pathological states D.sub.2-D.sub.4 are called parent pathological states, or parent first pathological states, in so far as they are directly linked to pathological state D.sub.1. They form pathological states of order two. The pathological state D.sub.2 (syndromic anomaly of the development of the eye) is the child of two parent pathological states D.sub.5 et D.sub.6, corresponding respectively to an anomaly of the development of the eye of genetic origin and an anomaly of the eye of genetic origin. These pathological states constitute pathological states of order three vis-à-vis the pathological state D.sub.1. Similarly, the pathological state D.sub.4, (otorhinolaryngological malformation), is the child pathological state of the pathological states D.sub.7 (otorhinolaryngological disease) to D.sub.Q (malformation of the head and neck). In a similar manner to the relationships between the signs and the pathological states, the logical relationships between the pathological states can comprise coefficients representative of a probability of kinship.

(14) FIG. 5 represents an example of a method for assistance with the establishment of a diagnosis of a patient according to the invention. The method 500 comprises a first step 501, in which one or more signs S.sub.i are entered by an operator into a device implementing the method, such as a computer. These signs can typically have been identified or assumed by an operator. They are thus called “identified signs”.

(15) In subsequent steps, potential pathological states of order one to N are sought. To this end, in a second step 502 a variable n takes the value one.

(16) In a subsequent step 503, all of the potential pathological states D.sub.j of order n are sought in the “pathological state” class. During the first iteration of this step 503, where n has the value one, the search is carried out by means of the first set 124 of logical relationships. During subsequent iterations of step 503, where n takes an integer value comprised between two and N, the search is carried out by means of the second set 125 of logical relationships. Step 503 of searching for potential pathological states D.sub.j results in a list of potential pathological states of different orders. The information relating to the order of each pathological state can be retained, with the objective of hierarchization, as explained below.

(17) In a fourth step 504, all of the potential signs S.sub.i of order n indicative of a potential pathological state D.sub.j of order n are identified in the signs class. This step 504 is carried out by means of the first set 124 of logical relationships. The step 504 can comprise a classification of the identified potential signs S, as a function of the order of the potential pathological state D.sub.j which has revealed them. Advantageously, when a potential sign S.sub.i is identified for different orders, it is possible to retain only the potential sign of the lowest order.

(18) In a step 505, all of the technical methods for acquisition of the medical images M.sub.k making it possible to observe the potential signs S.sub.i of order n are identified. More precisely, for each potential sign S.sub.i of order n, all the technical methods for acquisition of the medical images M.sub.k are identified in the sign class, by means of the third set 126 of logical relationships. These methods for acquisition are called technical methods for acquisition of interest. The step 505 can comprise a classification of the methods for acquisition of interest M.sub.k as a function of the order associated with the potential sign S.sub.i considered, so as to make it possible to classify these methods for acquisition M.sub.k according to whether they are more or less capable of showing potential signs S.sub.i close to the initial sign identified by the operator.

(19) In a step 506, it is verified whether the search for potential pathological states, and their associated potential signs, has sufficiently explored possible pathways. In practice it is verified whether the variable n is less than or equal to N, or strictly greater than N. In the first case, this variable is incremented by one unit in step a 507, and steps 502 to 506 are reiterated with this new value of n. In the second case, the search for new potential pathological states, associated new potential signs, and associated new methods for acquisition of interest ends.

(20) The method then executes a step 508 in which an examination protocol is constructed. This examination protocol consists of listing all of the technical methods for acquisition of interest M.sub.k, preferably as a function of their classification. The examination protocol thus constructed can be displayed to the operator in a step 509.

(21) In a step 510, the technical methods for acquisition of images to be implemented are selected from the technical methods of interest. The selection can be carried out by an operator, or automatically by an algorithm. It consists, for example, of selecting the methods for acquisition having the lowest order. Moreover, the methods for acquisition making it possible to observe the most potential signs over the same image, or involving the use of a single medical imaging device, can be selected to the detriment of methods for acquisition resulting in the observation of a single potential sign, or methods for acquisitions involving the use of different medical imaging devices.

(22) In an optional step 511, one or more reference images can be sought as a function of the methods for acquisition to be implemented. Step 511 consists of searching, in the library 11 of medical images, for those corresponding to the technical methods for acquisition to be implemented, and which thus represent potential signs. This search can be carried out via the label associated with each image. The operator having to carry out the acquisition of the medical images thus has reference images to allow them to verify that the acquired image corresponds to that expected for the subsequent establishment of a diagnosis.

(23) In a step 512, medical images are acquired. These medical images are acquired according to the technical methods for acquisition to be implemented. The acquisition of the images can be carried out by an operator, or automatically by one or more devices for acquisition of images.

(24) Of course, the invention is not limited to the examples which have just been described and numerous adjustments can be made to these examples without exceeding the scope of the invention. In particular, the signs of the ontology cannot necessarily be observed by means of one image, but can consist of any type of phenomenon linked to the state of a patient, such as the patient's body temperature, or a physical or biochemical property of a body fluid of the patient. Thus, the step of identifying technical methods for acquisition of medical images can consist, more generally, of a step of identifying technical methods for acquisition of medical data. Furthermore, the different characteristics, forms, variants and embodiments of the invention can be combined together according to various combinations to the extent that they are not incompatible or mutually exclusive.