The present invention proposes a data densifier comprising a plurality of first operation units and a second operation unit. The plurality of first operation units are respectively configured to be instantiated according to classification information, and the plurality of instantiated first operation units are configured to densify a plurality of sub data included in input data into a plurality of sub densified data. The second operation unit is configured to be instantiated according to the classification information, and the instantiated second operation unit is configured to merge the plurality of sub densified data from the plurality of first operation units into densified data. In addition, a data densification method used by the data densifier and a sense chip comprising the data densifier are also proposed.

A method and system for training a neural network are provided. The method includes receiving an input image, selecting at least one data augmentation method from a pool of data augmentation methods, generating an augmented image by applying the selected at least one data augmentation method to the input image, and generating a mixed image from the input image and the augmented image.

Various of the disclosed embodiments relate to systems and methods for processing surgical data to facilitate further downstream operations. For example, some embodiments may include machine learning systems trained to recognize whether video from surgical visualization tools, such as endoscopes, depicts a field of view inside or outside the patient body. The system may excise or whiteout frames of video appearing outside the patient so as to remove potentially compromising personal information, such as the identities of members of the surgical team, the patients identity, configurations of the surgical theater, etc. Appropriate removal of such non-surgical data may facilitate downstream processing, e.g., by complying with regulatory requirements as well as by removing extraneous data potentially inimical to further downstream processing, such as training a downstream classifier.

An image classifying device is provided in the invention. The image classifying device includes a storage device, a calculation circuit and a classifying circuit. The storage device stores information corresponding to a plurality of image classes. The calculation circuit obtains a target image from an image extracting device and obtains the feature vector of the target image. The calculation circuit obtains a first estimation result corresponding to the target image based on the information corresponding to the plurality of image classes and the feature vector and obtains a second estimation result corresponding to the target image based on a reference image, wherein the reference image corresponds to one of the image classes. The classifying circuit adds the target image into one of the image classes based on the first estimation result and the second estimation result.

An apparatus for identifying the state of an object inputs time series images into a first classifier to detect an object region including a predetermined object from each image, determines whether the region of each image is in a mixed state in which the region includes another object other than the object, chronologically inputs characteristics obtained from pixel values of the region of each image into a second classifier having a recursive structure, and applies a recursively used internal state of the second classifier stored in a memory to the second classifier, identifying the state of the object involving time-varying changes in outward appearance. The apparatus rejects the latest internal state when the region of each image is in the mixed state. The apparatus updates the internal state stored in the memory with this latest internal state when the region is not in the mixed state.

Methods and systems are described herein for hosting and arbitrating algorithms for the generation of structured frames of data from one or more sources of unstructured input frames. A plurality of frames may be received from a recording device and a plurality of object types to be recognized in the plurality of frames may be determined. A determination may be made of multiple machine learning models for recognizing the object types. The frames may be sequentially input into the machine learning models to obtain a plurality of sets of objects from the plurality of machine learning models and object indicators may be received from those machine learning models. A set of composite frames with the plurality of indicators corresponding to the plurality of objects may be generated, and an output stream may be generated including the set of composite frames to be played back in chronological order.

A system comprising: processors and memory containing instructions to control processors to: receive images representing an interior of a physical environment, identify, using neural network for object recognition, an object in an image, the object is associated with a location relative to the physical environment, identify, using neural network for object recognition, another object in another image, determine if objects in the images are located near or at a similar location based on location information associated with the objects, if the objects are located near or at a similar location, then objects are an instance of a single object, store similar location associated with the single object, display an interactive walkthrough visualization of a 3D model of the physical environment including the single object, receive request regarding object location through the interactive walkthrough visualization, and provide the similar location of the single object for display in the interactive walkthrough visualization.

Novel tools and techniques are provided for implementing digital microscopy imaging using deep learning-based segmentation and/or implementing instance segmentation based on partial annotations. In various embodiments, a computing system might receive first and second images, the first image comprising a field of view of a biological sample, while the second image comprises labeling of objects of interest in the biological sample. The computing system might encode, using an encoder, the second image to generate third and fourth encoded images (different from each other) that comprise proximity scores or maps. The computing system might train an AI system to predict objects of interest based at least in part on the third and fourth encoded images. The computing system might generate (using regression) and decode (using a decoder) two or more images based on a new image of a biological sample to predict labeling of objects in the new image.

A cervical cancer screening support system includes: an image acquisition unit that acquires a micrograph of a cell for cytodiagnosis of a cervix of uterus; a cell aggregate recognition unit that recognizes a cell aggregate in the micrograph; and an output unit that outputs a class applicable to a cell belonging to the cell aggregate.

Provided is a technique for determining a context of an image and an object depicted by the image based on the context. A trained context classification model may determine a context of an image, and a trained object recognition model may determine an object depicted by the image based on the image and the context. Provided is also a technique for determining an object depicted within an image based on an input location of an input detected by a display screen. An object depicted within an image may be detected based on a distance in feature space between an image feature vector of an image and a feature vector of the object, and a distance in pixel-space between an input location of an input and location of the object within the image.