Challenges of gated myocardial perfusion SPECT processing

Ploskikh, Victor; Kotina, E. D.

doi:10.35470/2226-4116-2021-10-3-171-177

Cited by 4 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An application of the described algorithm can be useful for different image processing fields, such as movement detection and its correction, tracing movement trajectory, constructing contours on an image, analyzing images and so on. Given approach is of a particular interest for processing radionuclide research data [Ovsyannikov, Kotina and Shirokolobov, 2013;Ploskikh and Kotina, 2021;Kotina, Ploskikh and Shirokolobov, 2022], for obtaining visual as well as quantitative information when diagnostic images are analyzed.…”

Section: Discussionmentioning

confidence: 99%

Optimization approach to the velocity field determining problem

Kotina

Ovsyannikov

Elizarova

2022

CAP

View full text Add to dashboard Cite

A reconstruction of the velocity field is of current interest for different theoretical and practical problems. In the presented work the optimization method of the velocity field construction for successive images is developed and generalized using the concepts of trajectory beam (an ensemble of trajectories), a distribution density function (brightness), and a quality functional. A new formulation of the problem is proposed with the splitting of the initial image to subsets and taking into account a temporal variation. Analytical expressions of the variation and gradient of the investigated functional are given, allowing one to construct various directed optimization methods. The proposed approach can be used for determining the velocity field within different tasks, particularly, for the problems of diagnostic image processing.

show abstract

Section: Discussionmentioning

confidence: 99%

Optimization approach to the velocity field determining problem

Kotina

Ovsyannikov

Elizarova

2022

CAP

View full text Add to dashboard Cite

show abstract

“…Currently, a large number of various modality images are used in medical diagnostics, which require processing and analysis: segmentation [Shen et al, 2023], detection and classification [Salemi et al, 2022], calculation of diagnostically significant parameters [Ploskikh and Kotina, 2021]. Deep learning approaches are widely used for these purposes [Liu et al, 2021].…”

Section: Introductionmentioning

confidence: 99%

Deep learning muscle segmentation model for CT images in DICOM format

Schmidt,

Kotina,

Buev

2023

CAP

View full text Add to dashboard Cite

This work solves the problem of automatic segmentation of medical images in DICOM format using machine learning methods. A new developed tool is used in the form of a separate module for labeling medical data in the DICOM format. The trained model, proposed in the paper, can be useful in the tasks of muscle segmentation. One can apply it in different ways, but some of the most common include assessment of diseases related to muscles, and sarcopenia is one of them. The further applications of the muscle segmentation model may include examining various medical cases with patients, that tend to have muscle-related diseases. For instance, detecting cachexia may be one of the extensions of the model’s application field.

show abstract

Virtual platform for computer simulation of radionuclide imaging in nuclear cardiology: Comparison with clinical data

Denisova,

Gurko,

Kolinko

et al. 2023

Digital Diagnostics

View full text Add to dashboard Cite

Background: In the field of radionuclide imaging, in vivo human clinical studies are limited due to radiation exposure and ethical concerns, so mathematical modeling and in silico computer simulations based on digital models are becoming increasingly important. In the English-language literature, this approach is called "Virtual clinical trials". Aims: The aim of this work is to develop software tools for simulation of radionuclide visualization of myocardial perfusion by SPECT/CT with 99mTc-MIBI and to perform studies aimed at improving the accuracy of SPECT. Materials and methods: A software package "Virtual platform for simulations of the SPECT/CT method in nuclear cardiology" was developed using digital patient models, a scanner and assessment of the state of the myocardium using digital images of the left ventricle (LV) in the form of a "polar map". Verification of the software package was performed by comparing with clinical data obtained at the National Medical Research Center of Cardiology (Moscow). Simulation computer tests were carried out, in which the accuracy of assessing the state of the myocardium was studied, depending on the approach to normalizing the "polar map" and corrective factors in the reconstruction algorithm. Results: The results of simulation tests showed that the assessment of LV myocardial perfusion significantly depends on the method of normalizing the "polar map" and taking into account corrective factors in the reconstruction algorithm. The most accurate estimates were obtained when calculating the normalization coefficient from the average value of activity in the normal zone of the myocardium. It is shown that the common approach to pixel normalization with maximum intensity can lead to errors. The results of "virtual" trials were fully consistent with clinical observations. Conclusions: The transition from relative normalized values of activity in the myocardium to absolute quantitative estimates will allow removing the existing limitations and uncertainties and is the main condition for improving the diagnostic accuracy of the SPECT/CT method in nuclear cardiology.

show abstract

Challenges of gated myocardial perfusion SPECT processing

Cited by 4 publications

References 16 publications

Optimization approach to the velocity field determining problem

Optimization approach to the velocity field determining problem

Deep learning muscle segmentation model for CT images in DICOM format

Virtual platform for computer simulation of radionuclide imaging in nuclear cardiology: Comparison with clinical data

Contact Info

Product

Resources

About