To provide an image processing apparatus, an image processing method, and an endoscope system by which deep tissues and superficial blood vessels enhanced with improved visibility can be displayed to a surgeon in real time. An image processing apparatus according to an embodiment of the present technology includes an enhancement processing unit that performs enhancement processing on low-frequency components that are a range lower than a predetermined spatial frequency in an input image, performs enhancement processing on high-frequency components that are a range higher than the low-frequency components in the input image, and outputs the input image having the low-frequency components and the high-frequency components each subjected to enhancement processing.

A prediction algorithm determines synthetic fluorescence images on the basis of measurement images. A validation of the synthetic fluorescence images can be effected on the basis of reference images which are captured after the measurement images or are captured for a separate sample. Alternatively or additionally, a training of the prediction algorithm can be effected on the basis of training images which are captured after the measurement images or are captured for a separate sample.

Disclosed is a fluorescence image analyzer for measuring and analyzing a sample that includes a plurality of cells in which target portions are labeled with fluorescent dyes, and the fluorescence image analyzer includes a light source configured to apply light to the sample; an imaging unit configured to take a fluorescence image of each of the cells by which fluorescence is generated by applying the light; a processing unit configured to process the fluorescence image having been taken; and a display unit. The processing unit obtains a bright point pattern of fluorescence in the fluorescence image, causes the display unit to display a plurality of positive patterns that are previously associated with at least one of a measurement item or a labeling reagent, and causes the display unit to display information of at least one of the number of abnormal cells included in the sample, a proportion of the number of the abnormal cells, the number of normal cells included in the sample, and a proportion of the normal cells, based on the bright point pattern having been obtained and the plurality of positive patterns.

Provided is a method for analyzing a tissue specimen, including treating a tissue specimen with an aqueous clearing agent and with at least two fluorescent probes to obtain a cleared and labeled tissue specimen; imaging the cleared and labeled tissue specimen to generate a three-dimensional (3D) image of the tissue specimen; preparing a stained tissue section from the cleared and labeled tissue specimen; capturing a reference two-dimensional (2D) image of the stained tissue section; matching the reference 2D image with the 3D image to extract from the 3D image a series of 2D image slices including a corresponding 2D image slice that corresponds to the reference 2D image; and determining at least one pathological score for each of the series of 2D image slices and reporting the presence or absence and the extent of the disease based on the pathological scores. The method can improve the accuracy of histopathologic diagnosis.

Apparatus and methods are described including placing at least a portion of a blood sample within a sample chamber (52), and acquiring microscopic images of the portion of the blood sample. Candidates of a given entity within the blood sample are identified, within the microscopic image. At least some of the candidates as being the given entity are validated, by performing further analysis of the candidates. A count of the candidates of the given entity is compared to a count of the validated candidates of the given entity, and at least the portion of the sample is invalidated from being used for performing at least some measurements upon the sample, at least partially based upon a relationship between the count of candidates and the count of validated candidates. Other applications are also described.

Image rotation in an endoscopic hyperspectral, fluorescence, and/or laser mapping imaging system is described. A system includes an emitter for emitting pulses of electromagnetic radiation and an image sensor comprising a pixel array for sensing reflected electromagnetic radiation. The system includes a rotation sensor for detecting an angle of rotation of a lumen relative to a handpiece of an endoscope. The system is such that at least a portion of the pulses of electromagnetic radiation emitted by the emitter comprises one or more of a hyperspectral emission, a fluorescence emission, and/or a laser mapping pattern.

According to an embodiment of the present invention, there is provided a method of screening a prostate cancer patient by optical image analysis of a circulating tumor cell marker and a prostate-specific membrane antigen.

The present disclosure relates generally to medical imaging, and more specifically to enhancing medical images (e.g., images taken in low-light conditions) using machine-learning techniques. An exemplary method of obtaining an enhanced fluorescence medical image of a subject comprises: receiving a fluorescence medical image of the subject (e.g., NIR images); providing the fluorescence medical image to a generator of a trained generative adversarial network (GAN) model trained using a plurality of white light images; and obtaining, from the generator, the enhanced fluorescence medical image of the subject.

Exemplary embodiments provide methods, mediums, and systems for processing multiplexed image data from a fluorescence in-situ hybridization (FISH) experiment. According to exemplary embodiments, a convolutional neural network (CNN) may be applied to the image data to localize and identify hybridization spots in images corresponding to different sets of targeting probes. The CNN is configured in such a way that it is able to discriminate hybridization spots in situations that are difficult for conventional techniques. The CNN may be trained on a relatively small amount of data by exploiting the nature of the FISH codebook.

A console of a mobile radiography apparatus includes a CPU that acquires a fluoroscopic image captured by a radiation detector and a visible light image captured by a visible light camera as a moving image related to the capture of the fluoroscopic image of a subject by the mobile radiography apparatus. The CPU extracts a frame to be subjected to a support process, which is a diagnosis support process or an imaging support process, from the moving image. A GPU executes the support process for the extracted frame.