Image Segmentation for Cardiovascular Biomedical Applications at Different Scales

Hemodynamics of vascular shunts: trends, challenges, and prospects

Vascular bypass surgery takes a significant place in the treatment of vascular disease. According to various assessments, this type of surgery is associated with almost 20 % of all vascular surgery episodes (up to 23 % according to the Federal Neurosurgical Center of Novosibirsk). Even though the problem of using of vascular grafts is obvious and natural, many problems associated with them are not still elucidated. From the mechanics' point of view, a vascular bypass is a converging or diverging tee, and the functioning of such structures still does not have strict mathematical formulations and proofs in the general case, which forces many researchers to solve specific engineering problems associated with shunting. Mathematical modeling, which is the gold standard for virtual simulations of industrial and medical problems, faces great difficulties and limitations in solving problems for vascular bypasses. Complications in the treatment of the vascular disease may follow the difficulties in mathematical modeling, and the price can be a cardiac arrest or a stroke. This work is devoted to the main aspects of the medical application of vascular bypasses and their functioning as a mechanical system, as well the mathematical aspects of their possible setup.

Coronary Centerline Extraction from CCTA Using 3D-UNet

The mesh-type coronary model, obtained from three-dimensional reconstruction using the sequence of images produced by computed tomography (CT), can be used to obtain useful diagnostic information, such as extracting the projection of the lumen (planar development along an artery). In this paper, we have focused on automated coronary centerline extraction from cardiac computed tomography angiography (CCTA) proposing a 3D version of U-Net architecture, trained with a novel loss function and with augmented patches. We have obtained promising results for accuracy (between 90–95%) and overlap (between 90–94%) with various network training configurations on the data from the Rotterdam Coronary Artery Centerline Extraction benchmark. We have also demonstrated the ability of the proposed network to learn despite the huge class imbalance and sparse annotation present in the training data.

Automated segmentation of abdominal organs from contrast-enhanced computed tomography using analysis of texture features

Generation of three-dimensional personalized geometric models of anatomical structures is an important process for many practical tasks: computer-aided diagnosis, treatment planning and numerical modeling in biomedical applications. Despite many efforts done by different research groups, automatic segmentation of organs still does not have any general solution. The main difficulties are caused by peculiarities of different medical imaging modalities, image variability (for the same modality) resulting from the wide range of imaging devices, noise and artifacts, large patient anatomical variability and overlapping of intensity ranges of neighboring anatomical structures. In this article, we propose segmentation method based on analysis of texture features and developed specially for segmentation of abdominal organs. Its main advantage is robustness to interpatient gray level and anatomical variability. The proposed method was validated on the patient data. The method implementation was accelerated using graphics processing unit (GPU).

Decision Support Software for Forecasting Patient’s Length of Stay

Length of stay of hospitalized patients is generally considered to be a significant and critical factor for healthcare policy planning which consequently affects the hospital management plan and resources. Its reliable prediction in the preadmission stage could further assist in identifying abnormality or potential medical risks to trigger additional attention for individual cases. Recently, data mining and machine learning constitute significant tools in the healthcare domain. In this work, we introduce a new decision support software for the accurate prediction of hospitalized patients’ length of stay which incorporates a novel two-level classification algorithm. Our numerical experiments indicate that the proposed algorithm exhibits better classification performance than any examined single learning algorithm. The proposed software was developed to provide assistance to the hospital management and strengthen the service system by offering customized assistance according to patients’ predicted hospitalization time.