Abstract: Techniques for obtaining a synthetic histochemically stained image from a multiplexed immunofluorescence (MPX) image may include producing an N-channel input image that is based on information from each of M channels of an MPX image of a tissue section, where M and N are positive integers and N is less than or equal to M; and generating a synthetic image by processing the N-channel input image using a generator network, the generator network having been trained using a training data set that includes a plurality of pairs of images. The synthetic image depicts a tissue section stained with at least one histochemical stain. Each pair of images of the plurality of pairs of images includes an N-channel image, produced from an MPX image of a first section of a tissue, and an image of a second section of the tissue stained with the at least one histochemical stain.

Abstract: The present disclosure relates to techniques for obtaining a synthetic immunohistochemistry (IHC) image from a histochemically stained image. Particularly, aspects of the present disclosure are directed to accessing an input image that depicts a tissue section that has been stained with at least one histochemical stain; generating a synthetic image by processing the input image using a trained generator network; and outputting the synthetic image. The synthetic image depicts a tissue section that has been stained with at least one IHC stain that targets a first antigen, and techniques may also include receiving an input that is based on a level of expression of a first antigen from the synthetic image and/or generating, from the synthetic image, a value that is based on a level of expression of the first antigen.

Abstract: Embodiments disclosed herein generally relate to identifying auto-fluorescent artifacts in a multiplexed immunofluorescent image. Particularly, aspects of the present disclosure are directed to accessing a multiplexed immunofluorescent image of a slice of specimen, wherein the multiplexed immunofluorescent image comprises one or more auto-fluorescent artifacts, processing the multiplexed immunofluorescent image using a machine-learning model, wherein an output of the processing corresponds to a prediction that the multiplexed immunofluorescent image includes one or more auto-fluorescent artifacts at one or more particular portions of the multiplexed immunofluorescent image, adjusting subsequent image processing based on the prediction, performing the subsequent image processing, and outputting a result of the subsequent image processing, wherein the result corresponds to a predicted characterization of the specimen.

Abstract: The present disclosure relates to techniques for transforming digital pathology images obtained by different slide scanners into a common format for image analysis. Particularly, aspects of the present disclosure are directed to obtaining a source image of a biological specimen, the source image is generated from a first type of scanner, inputting into a generator model a randomly generated noise vector and a latent feature vector from the source image as input data, generating, by the generator model, a new image based on the input data, inputting into a discriminator model the new image, generating, by the discriminator model, a probability for the new image being authentic or fake, determining whether the new image is authentic or fake based on the generated probability, and outputting the new image when the image is authentic.

Abstract: A multiplex image is accessed that depicts a particular slice of a particular sample stained with two or more dyes. Using a Generator network, a predicted singleplex image is generated that depicts the particular slice of the particular sample stained with each of the expressing biomarkers. The Generator network may have been trained by training a machine-learning model using a set of training multiplex images and a set of training singleplex images. Each of the set of training multiplex images depicted a slice of a sample stained with two or more dyes. Each of the set of training singleplex images depicted a slice of a sample stained with a single dye. The machine-learning model included a Discriminator network configured to discriminate whether a given image was generated by the Generator network or was a singleplex image of a real slide. The method further includes outputs the predicted singleplex image.

Abstract: The present disclosure relates machine learning techniques for segmenting non-tumor regions in specimen images to support tumor detection and analysis. Particularly, aspects of the present disclosure are directed to accessing one or more images that comprise a non-target region (e.g., a non-tumor region) and a target region (e.g., a tumor region), predicting, by a two-dimensional segmentation model, segmentation maps for the non-target region based on discriminative features encoded from the one or more images, a segmentation mask for the one or more images based on the segmentation maps, applying the segmentation mask to the one or more images to generate non-target region masked images that exclude the non-target region from the one or more images, and classifying, by an image analysis model, a biological material or structure within the target region based on a set of features extracted from the non-target region masked images.

Abstract: A medical imaging system includes a view transformation component (210) and a segment combiner (212). The transformation component (210) transforms projection data in each view of a plurality of individual segments, which each includes at least one view. The transformed projection data for substantially similar views across the plurality of individual segments have a common radius of rotation. The segment combiner (212) combines the transformed projection data to produce a single data set that includes the transformed projection data for each of the views of each of the plurality of individual segments.

Abstract: A presentation component (118) presents information from one or more data source(s) (102) to an assessor for assessment. A receiver operating characteristic (ROC) analyzer 120 uses an ROC analysis technique to evaluate the performance of the assessor. A feedback component (126) provides feedback as to the assessor's performance. A data manipulator (114) facilitates manipulation of the presented data.

Abstract: An imaging system (10) comprises at least one radiation detector (20) disposed adjacent a subject receiving aperture (18) to detect radiation from a subject, receive the radiation and generate measured data. An image processor (38) iteratively reconstructs the detected radiation into image representations, in each reconstruction iteration the image processor (38) applies noise reduction algorithms to at least a difference between the measured data and a portion of a previous iteration image representation.

Abstract: A potential region of interest segmentation device segments an image data into regions of potential interest. From the regions of potential interest, an identifying device identifies a region of interest including an object of interest based on a set of rules. A recognizing device differentiates among the identified objects of interest and selects at least one particular object of interest.

Abstract: A presentation component (118) presents information from one or more data source(s) (102) to an assessor for assessment. A receiver operating characteristic (ROC) analyzer 120 uses an ROC analysis technique to evaluate the performance of the assessor. A feedback component (126) provides feedback as to the assessor's performance. A data manipulator (114) facilitates manipulation of the presented data.

Abstract: A medical imaging system (10) includes at least one radiation detection head (16) disposed adjacent a subject receiving aperture (18) to detect radiation from a subject. The detected radiation is reconstructed into at least one initial 2D projection image (?). Resolution in each initial 2D image (?) is restored by using the extended iterative constrained deconvolution algorithm by incorporating different estimates of the system response function which estimates correspond to different distances between the detection head and the origins of the detected radiation. Measured response functions are used to restore a series of images. The optimal image is determined by automatic searching with the figure of merit, by user's observation, or by using blind deconvolution for a concurrent estimating of the system response function and updating the original image.

Abstract: A method includes obtaining a combined data set that includes first and second imaging data sets. The first and second imaging data sets correspond to different imaging modalities. The method further includes determining a metric indicative of an alignment between the first and second imaging data sets in the combined data set. The method further includes presenting the metric in a human readable format.

Abstract: A medical imaging system includes a view transformation component (210) and a segment combiner (212). The transformation component (210) transforms projection data in each view of a plurality of individual segments, which each includes at least one view. The transformed projection data for substantially similar views across the plurality of individual segments have a common radius of rotation. The segment combiner (212) combines the transformed projection data to produce a single data set that includes the transformed projection data for each of the views of each of the plurality of individual segments.

Abstract: When estimating a position or orientation of a patient's heart, a mesh model of a nominal heart is overlaid on a SPECT or PET image of the patient's heart and manipulated to conform to the image of the patient's heart. A mesh adaptation protocol applies opposing forces to the mesh model to constrain the mesh model from changing shape and to pull the mesh model to the shape of the patient's heart. A heart orientation estimator (60) iterates the mesh adaptation protocol a predetermined number of times, after which it defines a long axis of the left ventricle of the patient's heart as a line passing through the center of the mitral valve and the center of mass of the left ventricle. The long axis is then employed by a reorientation processor (70) to reorient the SPECT or PET image of the patient's heart, over which the mesh model was originally laid, to improve the accuracy of the PECT or PET image.

Abstract: A phantom (44) is used to calibrate a multi-modality imaging system (10) that includes a nuclear imaging system (12) and a CT scanner (14). The phantom (44) includes marker receiving cavities (96) positioned at fixed locations in the phantom, in which markers (46) are removably placed. The markers (46) include CT markers (90), which are imageable by the CT scanner (14), and radioisotope markers (48), which are imageable by the nuclear imaging system (12). The radioisotope markers (48) are disposed into wells (92) provided at a center of mass of each disk-like CT marker. The markers (46) have a label (94) identifying its isotope. The phantom (44), rigidly affixed to a couch (32), is imaged by the nuclear imaging system (12) and by the CT scanner (14). A transformation processor (72) calculates a transformation which brings centroids of the CT markers (90) in a CT image and the radioisotope markers (48) in a nuclear image into alignment.

Abstract: An imaging system (10) comprises at least one radiation detector (20) disposed adjacent a subject receiving aperture (18) to detect radiation from a subject, receive the radiation and generate measured data. An image processor (38) iteratively reconstructs the detected radiation into image representations, in each reconstruction iteration the image processor (38) applies noise reduction algorithms to at least a difference between the measured data and a portion of a previous iteration image representation.

Abstract: A medical imaging system (10) includes at least one radiation detection head (16) disposed adjacent a subject receiving aperture (18) to detect radiation from a subject. The detected radiation is reconstructed into at least one initial 2D projection image (?). Resolution in each initial 2D image (?) is restored by using the extended iterative constrained deconvolution algorithm by incorporating different estimates of the system response function which estimates correspond to different distances between the detection head and the origins of the detected radiation. Measured response functions are used to restore a series of images. The optimal image is determined by automatic searching with the figure of merit, by user's observation, or by using blind deconvolution for a concurrent estimating of the system response function and updating the original image.

Abstract: A phantom (44) is used to calibrate a multi-modality imaging system (10) that includes a nuclear imaging system (12) and a CT scanner (14). The phantom (44) includes marker receiving cavities (96) positioned at fixed locations in the phantom, in which markers (46) are removably placed. The markers (46) include CT markers (90), which are imageable by the CT scanner (14), and radioisotope markers (48), which are imageable by the nuclear imaging system (12). The radioisotope markers (48) are disposed into wells (92) provided at a center of mass of each disk-like CT marker. The markers (46) have a label (94) identifying its isotope. The phantom (44), rigidly affixed to a couch (32), is imaged by the nuclear imaging system (12) and by the CT scanner (14). A transformation processor (72) calculates a transformation which brings centroids of the CT markers (90) in a CT image and the radioisotope markers (48) in a nuclear image into alignment.

Abstract: A potential region of interest segmentation device segments an image data into regions of potential interest. From the regions of potential interest, an identifying device identifies a region of interest including an object of interest based on a set of rules. A recognizing device differentiates among the identified objects of interest and selects at least one particular object of interest.