A system and computation method is disclosed that identifies radiation-induced lung injury after radiation therapy using 4D computed tomography (CT) scans. After deformable image registration, the method segments lung fields, extracts functional and textural features, and classifies lung tissues. The deformable registration locally aligns consecutive phases of the respiratory cycle using gradient descent minimization of the conventional dissimilarity metric. Then an adaptive shape prior, a first-order intensity model, and a second-order lung tissues homogeneity descriptor are integrated to segment the lung fields. In addition to common lung functionality features, such as ventilation and elasticity, specific regional textural features are estimated by modeling the segmented images as samples of a novel 7.sup.th-order contrast-offset-invariant Markov-Gibbs random field (MGRF). Finally, a tissue classifier is applied to distinguish between the injured and normal lung tissues.

An information processing apparatus includes a determination unit configured to determine, as a threshold value to which a signal intensity value corresponding to a shadow region is compared, a value different from a threshold value to which a signal intensity value corresponding to a region other than the shadow region is compared.

The present invention relates to deep learning for automated segmentation of a medical image. More particularly, the present invention relates to deep learning for automated segmentation of anatomical regions and lesions in mammography screening and clinical assessment.

According to a first aspect, there is provided a computer-aided method of segmenting regions in medical images, the method comprising the steps of: receiving input data; analysing the input data by identifying one or more regions; determining one or more characteristics for the one or more regions in the input data; and generating output segmentation data in dependence upon the characteristics for the one or more regions.

A system and method for automated characterization of solid tumors using medical imaging. The system comprises an interface that is configured to acquire data from medical imaging devices, one or more processors, and an outputting device that reports the characterization of said solid tumor. The method of automated characterization, which is implemented by the system, acquires a sequence of images from the medical imager using a Dynamic Contrast Enhanced (DCE) imaging protocol, performs image registration, detects the contour of the solid tumor, and dividing the contours to segments. For each segment, the method calculating a displacement of the contrast material, fitting the displacement to a flow model and extracting an estimation of the interstitial fluid velocity. The estimated interstitial fluid velocity of the segments provide characterization of the solid tumor and includes an assessment of the tumor interstitial fluid pressure, the tumor drug delivery efficiency, and the tumor prognostic or metastasis risk.

Methods are presented for simulating drug movement within a model of a tumor that is mapped to the specific anatomy of the corresponding tumor in a patient as determined by imaging. With a segmentation of the tumor into distinct interconnected compartments and pre-determined initial parameters for the distributed tissue diffusivities and perfusion levels, the disclosed techniques can be used to predict the drug concentration throughout the tumor as a function of time, as well as the accumulation of drug in the rest of the body. In this way, advantageous initial parameters may be determined using the model. It is also possible to predict drug concentration throughout the tumor for a given intravenous injection of drug. Such a model serves an important role in treatment planning for lung tumors.

A method, an apparatus and a system for cell detection are provided. In the apparatus, a hyperspectrum module is used to capture information across electromagnetic spectrums from an image, a stereo camera module is used to capture three-dimensional image information, and the hyperspectrum module and the stereo camera module form a trinocular micro spectrometer. A microscopic optical module is provided for the two modules to form hyperspectrum and three-dimensional image information from a cell and its split cells via a lens. In the method, a series of continuous images are obtained within a time period. An observation image array with a plurality of observation image zones are provided to retrieve coordinates of a plurality of feature points at different times. Finally, a holistic cellular activity can be obtained by analyzing continuous hyperspectrum and 3D image information from the images over time.

An insole design method and an insole design system are provided, and the method includes: capturing an uncompressed free foot model by a depth camera and obtaining a free foot model three-dimensional image; capturing a pressed foot model stepped on a transparent pedal by the depth camera and obtaining a pressed foot model three-dimensional image; aligning the free foot model three-dimensional image with the pressed foot model three-dimensional image; calculating and obtaining a plantar deformation quantity according to the aligned free foot model three-dimensional image and the aligned pressed foot model three-dimensional image; and completing the designed insole according to a sole projection plane or a three-dimensional profile of the specific sole and the plantar deformation quantity.

In a related-art technology, training images in accordance with the number of classifications need to be prepared to perform training process. For this reason, in a case where a model is caused to output how much effect of a drug is expressed, a training image needs to be prepared for each expression degree, and it is troublesome to create the model. Provided is an apparatus including an image obtaining unit configured to obtain an evaluation target image depicting a subject of an evaluation target, a probability obtaining unit configured to obtain a recognition probability regarding the evaluation target image by using a model that outputs, in accordance with input of an image, a recognition probability at which a subject of the image is recognized as the subject before effect of a drug is expressed or the subject after the effect of the drug is expressed, and a calculation unit configured to calculate an expression degree of the effect of the drug based on the recognition probability of the evaluation target image.

A heart rate detection method includes a facial image data acquiring step, a feature points recognizing step, an effective displacement signal generating step and a heart rate determining step. The feature points recognizing step is for recognizing a plurality of feature points, wherein a number range of the feature points is from three to twenty, and the feature points include a center point between two medial canthi, a point of a pronasale and a point of a subnasale of the face. The effective displacement signal generating step is for calculating an original displacement signal, wherein the original displacement signal is converted to an effective displacement signal. The heart rate determining step is for transforming the effective displacement signals of each of the feature points to an effective spectrum, wherein a heart rate is determined from one of the effective spectrums corresponding to the feature points, respectively.

The present invention relates to apparatus for automatic quantification of a part of vascular structure. It is described to provide (12) at least one first image comprising a spatial representation of a region of interest of a vascular structure, wherein the at least one first image comprises image data representative of a location of a part of a medical device. The medical device is configured to be used in a vascular treatment, and the part of the medical device is configured to be in a plurality of states associated with different phases of the vascular treatment. At least one second image comprising a spatial representation of the region of interest of the vascular structure is provided (14), wherein the at least one second image comprises image data representative of at least a part of the vascular structure in a visible and distinct manner. A location of a feature in the spatial representation of the region of interest of the vascular structure of the at least one first image is determined (18), wherein the feature is associated with the part of the medical device in one of the states associated with a phase of the vascular treatment. A transform relating at least one location in the at least one first image to a corresponding at least one location in the at least one second image is determined (20) and applied to the location of the feature in the spatial representation of the region of interest of the vascular structure of the at least one first image to provide a determined location in the spatial representation of the region of interest of the vascular structure of the at least one second image. Data is output (22) representative of the vascular structure at the determined location.