57 results about "Reviewing Radiologist" patented technology

A marking device for defining the margins and orientation of radiography specimens includes a plurality of visually distinctive markers joined by a base that holds the markers until the markers are secured to a specimen. The base serves as a holder for the individual markers as the markers are secured to a specimen with sutures or staples, at which time the markers are disconnected from the base. One or more markers may be secured to a specimen as needed to define the orientation of the specimen. By securing markers to specific locations on a specimen, a surgeon can indicate to radiologists and pathologists the specimen's orientation in the body before removal, thus aiding in future study of the specimen. Since the markers are made either wholly or partially from radiopaque material, the markers are visible when radiographed.

The present invention provides for the use of computer-aided detection (CAD) system output displays for providing accurate representations of areas for subsequent exams. Since the CAD output, unlike the original medical imagery, is not used during the initial reading, the radiologist does not mark it until a final determination is reached regarding subsequent procedures. Additionally, since the CAD output contains versions of the original imagery, the regions indicated by the radiologist are shown in the context of the particular anatomical detail for a given patient. This detail assists the technologist, other physicians and patients in more efficiently and accurately locating the exact area for subsequent exams.

The business method known as QAISys (Quality Assessment and Improvement System) is a process that rates comprehensively and continually the quality of medical images as determined by the interpreting radiologist. The output of this system conveys feedback from the radiologist to the performing technologist and his/her supervisor. This feedback enables medical radiologists, technologists, and managers to bridge any communication gaps between themselves in respect to the quality and effectiveness of medical images. It permits management to assess and track image quality for an entire medical imaging department by modality, location, and/or shift, and for each individual technologist. The QAISys method reveals the nature of recurrent quality failures and highlights which exam types need to be improved. QAISys then indicates practical, cost-effective means of assessing and improving the overall medical images for future patients.

The invention discloses a deep residual network-based semantic mammary gland molybdenum target image lump segmentation method. The method comprises the following steps of: labelling pixel categories of lumps and normal tissues corresponding to a collected mammary gland molybdenum target image so as to generate label images, and dividing the mammary gland molybdenum target image and the corresponding label images into training samples and test samples; preprocessing the training samples to form a training data set; constructing a deep residual network, and training the network by utilizing thetraining data set, so as to obtain a deep residual network training model; after a to-be-segmented mammary gland molybdenum target image lump is preprocessed, carrying out binary classification and post-processing on a pixel of the to-be-segmented mammary gland molybdenum target image by utilizing the deep residual network training model, and outputting lump segmentation image to realize semanticsegmentation of the mammary gland molybdenum target image lump. The method is capable of effectively improving the automatic and intelligent levels of mammary gland molybdenum target image lump segmentation, and can be applied to the technical field of assisting radiologists to carry out medical diagnosis.

A method of detecting and diagnosing cancers characterized by the presence of at least one nodule/neoplasm from an imaging scan is presented. To detect nodules in an imaging scan, a 3D CNN using a single feed forward pass of a single network is used. After detection, risk stratification is performed using a supervised or an unsupervised deep learning method to assist in characterizing the detected nodule/neoplasm as benign or malignant. The supervised learning method relies on a 3D CNN used with transfer learning and a graph regularized sparse MTL to determine malignancy. The unsupervised learning method uses clustering to generate labels after which label proportions are used with a novel algorithm to classify malignancy. The method assists radiologists in improving detection rates of lung nodules to facilitate early detection and minimizing errors in diagnosis.

The invention discloses a prostatic cancer computer-assisted detection method and system based on multi-parameter MRI and relates to the field of medical image processing. The method includes the following steps that at a training stage, firstly, a clinical case sample is preprocessed, then a prostate region and a focus candidate region are automatically extracted, and then features of the focus candidate region are calculated and used for training a classifier; at a testing stage, the trained classier is used for classifying the features of the focus candidate region automatically extracted from the tested clinical case sample, and a corresponding diagnosis result is obtained and serves as reference comments to be provided for doctors. A series of quantitative indexes and corresponding malignancy probability values are provided for radiologists, and the doctors can be effectively assisted in diagnosing the prostatic cancer through an MRI image.

A system and method to increase the efficiency and effectiveness of a radiologist. A context manager for managing the display of multi-patient contexts. In an embodiment computer software operates as a context manager and correlates the context of at least one information container with at least one image container. A graphical container displaying information about a patient, as opposed to an image, may be an information container. A graphical container displaying images of a patient, as opposed to information, may be an image container. In order to minimize occurrences of a user viewing information about a first patient and images of a second patient, and mistakenly believing the information and images belong to the same patient, a visual indicator may be displayed over images that are out-of-context with the information displayed. The visual indicator may be a watermark to alert a user of a difference in context.

The invention presents a way to automatically assess the quality of acquired (S1, S2) digital mammographic images with respect to the image positioning of a patient's breast. The automated digital quality assessment (S3) is executed in real time and preferably notifies (S4) the technologist instantly if the image positioning quality of a mammographic image is inadequate. This makes it possible to retake the image while the patient is still present at the examination facility. The quality assessment notification includes information to the technologist, preferably both visually and statistically of land-mark positioning measurements. Alternatively, the digital quality assessment is accompanied by a computerized decision of whether the mammogram needs to be retaken, requiring a minimum of involvement, if at all, from the technologist. The invention hence provides a set of quality-assured mammographic images that can be stored (S5) and later accessed by a radiologist for review and diagnosis (S6).

A mammographic apparatus comprises an identification information reader and the identification information reader transmits the cassette identification information read by the identification information reader together with the radiography execution information to the control apparatus, in the control apparatus, it becomes possible to automatically correlate the radiographic order information and the radiography execution information based on the cassette ID (identification information). Accordingly, since the radiologist can eliminate the selection work to select of the radiographic order information for correlating the radiography execution information prior to the radiography, the efficiency of radiographic work can be improved. In the case of mammography which conducts a plurality of radiographs with different radiographic body parts with different radiographic directions, since the selection operation can be eliminated, the radiologist does not need to move the control apparatus for a selection operation and especially, the efficiency of radiographic work can be improved.

A physician in an emergency room inputs a radiology order into a first mobile device. A radiologist receives the order over a second mobile device and inputs a report into the device. The devices provide prompts to input the information according to American College of Radiology guidelines. Information summaries for billing are produced. Physicians who frequently input inappropriate orders can be identified. A learning system uses patient clinical outcomes and pathology results to assess the usefulness of the examinations being performed.

The invention discloses to a multi-parameter MRI prostate cancer CAD method and system based on two kinds of classifiers, relating to the medical image processing field. The multi-parameter MRI prostate cancer CAD method comprises candidate focus automatic detection and candidate focus computer aided diagnosis. The candidate focus automatic detection comprises steps of respectively performing pre-processing on three MRI sequences of each case: T2WI, DWI, ADC to make resolution ratios and sizes of the T2WI, the DWI, the ADC identical, wherein pixels of a same position basically correspond to a same part of a human body, and respectively extracting point characteristics on three kinds of MRI sequences of each case, and inputting the point characteristics into a focus detection classifier to obtain a candidate focus. The candidate focus computer aided diagnosis comprises steps of calculating regional characteristics of the candidate focus in three kinds MRI sequences of each case and inputting the regional characteristics into a focus diagnosis classifier to obtain a corresponding diagnosis result. The multi-parameter MRI prostate cancer CAD method and system based on two kinds of classifiers can provide a series of quantized indexes and a malignant probability value.

The invention discloses a cloud storage and medical image seamless joint system. The system comprises a medical image information acquisition device, a radiology department server, an imaging center server, a front server, a storage gateway, a cloud storage, a radiologist workstation, a clinician workstation, and a tape backup device, wherein the medical image information acquisition device is used for sending acquired information to the radiology department server and the information is copied through the tape backup device, the radiology department server is connected with the imaging center server and used for sending data to an imaging center, the imaging center server is connected with the front server and used for sending the data to the front server; the front server is used for sending the data to the storage gateway, and then the data is stored in the cloud storage through the storage gateway.

The invention relates to the field of medical image technology, and particularly to a breast nodules auxiliary diagnosis method based on DSSD and a system. The system comprises a breast image workstation, a breast cloud diagnosis system and a breast diagnosis mobile terminal. The breast image workstation performs breast image acquisition. The breast cloud diagnosis system comprise training set anda testing set preprocessing, marking and neural network discriminating, wherein the training set and the testing set preprocessing comprises multi-scale image de-noising and enhancing. The method comprises the steps of acquiring a breast graph by the breast image workstation, performing discriminating through mainly using a DSSD neural network, and finally transmitting a result which is obtainedthrough processing of the breast cloud diagnosis system to the breast diagnosis mobile terminal. According to the method and the system, a computer-aided diagnosis (CAD) system based on deep learningis developed for aiming at the field of the medical image (unnatural image), thereby supplying a second opinion for diagnosis to a radiology department doctor, and more effectively assisting the doctor in performing diagnosis.

Imaging protocols are modified in real time through implementation of artificial intelligence. Key steps include: inputting of medical images and patient data; performing artificial intelligence analysis; outputting a potentially modified radiological imaging examination protocol; if applicable, an option for the radiologist to review the patient data, images acquired and the AI's potentially modified radiological imaging examination protocol; delivery of the modified radiological imaging examination protocol to the imaging device; and, if applicable, an option for the radiologist to provide feedback and re-train the artificial intelligence algorithm. Further, some of the images that are utilized by the AI system include processed images, such as segmented and filtered volume rendered images.

The invention relates to a medical image efficient classification management method based on big data, and the method comprises the steps: carrying out the image standardization of a T2 weighted magnetic resonance image, and obtaining a standardized image; performing gaussian filtering on the standardized image; performing contrast stretching on the denoised image; extracting brightness features of the stretched image; carrying out histogram equalization on the stretched image to obtain an image with enhanced contrast; extracting texture features of the contrast-enhanced image through a gray level co-occurrence matrix; training and verifying the classification model through a support vector machine by adopting a one-leaving method; establishing a graphical user interface for human-computerinteraction. According to the invention, the classification precision of the classifier is improved and the time complexity of the algorithm is reduced through background removal; through two times of image enhancement, the texture of the image is more obvious, so that the classification precision is improved; the working efficiency of radiologists can be improved; different features are extracted for different stages, and image features of different stages are met.

A method and system for medical image pre-processing at the medical image scanner that facilitates joint interpretation of the medical images by radiologists and artificial intelligence algorithms is disclosed. Raw medical image data is acquired by performing a medical image scan of a patient using a medical image scanner. Input data associated with the medical image scan of the patient and available downstream automated image analysis algorithms is acquired. A set of pre-processing algorithms to apply to the raw medical image data is selected based on the input data associated with the medical image scan of the patient and the available downstream automated image analysis algorithms using a trained machine learning based model. One or more medical images are generated from the raw medical image data by applying the selected set of pre-processing algorithms to the raw medical image data.

An artificial intelligence engine for directed hypothesis generation and ranking uses multiple heterogeneous knowledge graphs integrating disease-specific multi-omic data specific to a patient or cohort of patients. The engine also uses a knowledge graph representation of ‘what the world knows’ in the relevant bio-medical subspace. The engine applies a hypothesis generation module, a semantic search analysis component to allow fast acquiring and construction of cohorts, as well as aggregating, summarizing, visualizing and returning ranked multi-omic alterations in terms of clinical actionability and degree of surprise for individual samples and cohorts. The engine also applies a moderator module that ranks and filters hypotheses, where the most promising hypothesis can be presented to domain experts (e.g., physicians, oncologists, pathologists, radiologists and researchers) for feedback. The engine also uses a continuous integration module that iteratively refines and updates entities and relationships and their representations to yield higher quality of hypothesis generation over time.