Andrey Popandopulo

Dr. Andrey Popandopulo


Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
Maksim SolopovMaksim Solopov
Maksim Solopov
ECElizaveta Chechekhina
Elizaveta Chechekhina
AKAnna Kavelina
Anna Kavelina
GAGulnara Akopian
Gulnara Akopian
VTViktor Vasilevich Turchin
Viktor Vasilevich Turchin
RIRoman Ishchenko
Roman Ishchenko
The aim of this study is to conduct a comparative assessment of the effectiveness of neural network models—U-Net, DeepLabV3+, SegNet and Mask R-CNN—for the semantic segmentation of micrographs of human mesenchymal stem cells (MSCs). A dataset of 320 cell micrographs annotated by cell biology experts was created. The models were trained using a transfer learning method based on ImageNet pre-trained weights. As a result, the U-Net model demonstrated the best segmentation accuracy according to the metrics of the Dice coefficient (0.876) and the Jaccard index (0.781). The DeepLabV3+ and Mask R-CNN models also showed high performance, although slightly lower than U-Net, while SegNet exhibited the least accurate results. The obtained data indicate that the U-Net model is the most suitable for automating the segmentation of MSC micrographs and can be recommended for use in biomedical laboratories to streamline the routine analysis of cell cultures.

18 October 2025 in Journal of Imaging
Verified
Journal Article
Open access
PDF file
Maksim SolopovMaksim Solopov
Maksim Solopov
ECElizaveta Chechekhina
Elizaveta Chechekhina
VTViktor Vasilevich Turchin
Viktor Vasilevich Turchin
APAndrey Popandopulo
Andrey Popandopulo
Dmitry FilimonovDmitry Filimonov
Dmitry Filimonov
RIRoman Ishchenko
Roman Ishchenko
This scoping review explores the application of artificial intelligence (AI) methods for analyzing mesenchymal stem cells (MSCs) images. The aim of this study was to identify key areas where AI-based image processing techniques are utilized for MSCs analysis, assess their effectiveness, and highlight existing challenges. A total of 25 studies published between 2014 and 2024 were selected from six databases (PubMed, Dimensions, Scopus, Google Scholar, eLibrary, and Cochrane) for this review. The findings demonstrate that machine learning algorithms outperform traditional methods in terms of accuracy (up to 97.5%), processing speed and noninvasive capabilities. Among AI methods, convolutional neural networks (CNNs) are the most widely employed, accounting for 64% of the studies reviewed. The primary applications of AI in MSCs image analysis include cell classification (20%), segmentation and counting (20%), differentiation assessment (32%), senescence analysis (12%), and other tasks (16%). The advantages of AI methods include automation of image analysis, elimination of subjective biases, and dynamic monitoring of live cells without the need for fixation and staining. However, significant challenges persist, such as the high heterogeneity of the MSCs population, the absence of standardized protocols for AI implementation, and limited availability of annotated datasets. To advance this field, future efforts should focus on developing interpretable and multimodal AI models, creating standardized validation frameworks and open-access datasets, and establishing clear regulatory pathways for clinical translation. Addressing these challenges is crucial for accelerating the adoption of AI in MSCs biomanufacturing and enhancing the efficacy of cell therapies.

8 August 2025 in Journal of Imaging
Verified
Journal Article
Open access
PDF file
RIRoman Ishchenko
Roman Ishchenko
Maksim SolopovMaksim Solopov
Maksim Solopov
APAndrey Popandopulo
Andrey Popandopulo
ECElizaveta Chechekhina
Elizaveta Chechekhina
VTViktor Vasilevich Turchin
Viktor Vasilevich Turchin
Dmitry FilimonovDmitry Filimonov
Dmitry Filimonov
This study evaluates the effectiveness of transfer learning with pre-trained convolutional neural networks (CNNs) for the automated binary classification of surgical suture quality (high-quality/low-quality) using photographs of three suture types: interrupted open vascular sutures (IOVS), continuous over-and-over open sutures (COOS), and interrupted laparoscopic sutures (ILS). To address the challenge of limited medical data, eight state-of-the-art CNN architectures—EfficientNetB0, ResNet50V2, MobileNetV3Large, VGG16, VGG19, InceptionV3, Xception, and DenseNet121—were trained and validated on small datasets (100–190 images per type) using 5-fold cross-validation. Performance was assessed using the F1-score, AUC-ROC, and a custom weighted stability-aware score (Scoreadj). The results demonstrate that transfer learning achieves robust classification (F1 > 0.90 for IOVS/ILS, 0.79 for COOS) despite data scarcity. ResNet50V2, DenseNet121, and Xception were more stable by Scoreadj, with ResNet50V2 achieving the highest AUC-ROC (0.959 ± 0.008) for IOVS internal view classification. GradCAM visualizations confirmed model focus on clinically relevant features (e.g., stitch uniformity, tissue apposition). These findings validate transfer learning as a powerful approach for developing objective, automated surgical skill assessment tools, reducing reliance on subjective expert evaluations while maintaining accuracy in resource-constrained settings.
1

6 November 2024 in Siberian journal of oncology
Verified
Journal Article
Open access
PDF file
Maksim SolopovMaksim Solopov
Maksim Solopov
AKAnna Kavelina
Anna Kavelina
APAndrey Popandopulo
Andrey Popandopulo
VTViktor Vasilevich Turchin
Viktor Vasilevich Turchin
SPS. A. Pashchenko
S. A. Pashchenko
KBK. M. Bagdasarov
K. M. Bagdasarov
Objective: to evaluate the effectiveness of a convolutional neural network model for automated cytologic diagnosis of papillary thyroid cancer and benign thyroid nodules.Material and Methods. The convolutional neural network was developed in the Python programming language using the TensorFlow 2.15.0 open source library. For the study, a dataset that included two categories of pathologies was generated: 1597 microphotographs of papillary carcinoma and 767 microphotographs of benign nodules (colloid goiter and adenomatous nodules). To form a training sample and evaluate the model’s performance metrics on the test sample, the dataset was divided in a ratio of 80/20.Results. In classifying papillary carcinoma, the model achieved precision of 89.3 %, recall of 92.4 %, specifcity of 77.4 % and F1 score of 91.4 %. When identifying benign nodules, the presicion, recall, specifcity and F1 score were 83.3 %, 77.4 %, 92.4 %, and 80.3 %, respectively, indicating a higher rate of false-positive and false-negative predictions. The AUC was 0.91 at the individual microphotograph level and 0.94 at the serial microphotograph level from one patient, indicating the high ability of the trained model to differentiate between malignant and benign thyroid lesions based on microphotographs of fne-needle aspiration biopsy specimens.Conclusion. Further improvement of the neural network model by training on larger and more diverse datasets of microphotographs of cytological specimens of the thyroid gland will help improve its diagnostic range and performance. The developed model can be used to develop software for identifying thyroid pathologies.

30 June 2022 in Patologiya krovoobrashcheniya i kardiokhirurgiya
Verified
Journal Article
Open access
PDF file
APAndrey Popandopulo
Andrey Popandopulo
Maksim SolopovMaksim Solopov
Maksim Solopov
VTViktor Vasilevich Turchin
Viktor Vasilevich Turchin
VBV.V. Bushe
V.V. Bushe
The high mortality rate from heart disease requires the development of new therapeutic approaches. Traditional treatments can provide a compensatory effect, improve the patient's quality of life, but not eliminate myocardial loss. Preclinical studies indicate that the regeneration of cardiac tissue can be stimulated through transplantation of stem and progenitor cells. The review describes clinical trials of myocardial injury therapy using regional stem cells, progenitor cells, and cells differentiated from ESC/iPSCs. The results of clinical studies have confirmed the safety and availability of cell therapy for myocardial injury. However, insufficient knowledge of the mechanisms of action of transplanted cells on the myocardium and the rapid deployment of clinical trials initiated by the success of preclinical trials made it possible to observe, at best, only a moderate therapeutic effect. In studies using regional stem cells, it was not possible to reliably confirm their ability to differentiate into cardiomyocytes and integrate into the myocardium, and the observed improvements in heart function are associated with paracrine signaling. Cardiac ESC/iPSC derivatives are more suitable for the restoration of the lost myocardium, but since the number of clinical trials with these types of cells is still small, additional studies are required to confirm their safety and efficacy. When developing the design of future clinical trials of cell therapy, it is necessary to strictly take into account the mechanism of cell action, the pathophysiology of the disease, and select the optimal set of endpoints. The experience of the conducted research leads to an understanding of the need to develop personalized cell therapy, according to which the choice of the source of cells and the route of administration should be based on the individual characteristics of the patient. The growing number of clinical trials and the active study of the mechanisms of action of transplanted cells on damaged myocardium suggest that this type of cell therapy has a chance to enter clinical practice in the near future. Received 4 October 2021. Revised 13 December 2021. Accepted 20 December 2021. Funding: The study did not have sponsorship. Conflict of interest: Authors declare no conflict of interest. Contribution of the authors: The authors contributed equally to this article.

30 August 2026 in Siberian Scientific Medical Journal
Verified
Journal Article
Open access
PDF file
Maksim SolopovMaksim Solopov
Maksim Solopov
APAndrey Popandopulo
Andrey Popandopulo
VTViktor Vasilevich Turchin
Viktor Vasilevich Turchin
AEA. A. Ermak
A. A. Ermak
SRS. A. Rudskoy
S. A. Rudskoy
РИР. В. Ищенко
Р. В. Ищенко
Accurate measurement of wound geometric parameters is essential for assessing injury severity and monitoring healing progress. Traditional manual methods have significant limitations: high error rates, subjectivity, and risk of infection. Modern digital techniques lack these drawbacks but are often expensive and complex. This study presents an alternative solution: an automated scanning platform assembled from affordable components.    Aim of the study  was to develop and validate a robotic XY-scanning platform for automated 3D analysis of ex vivo models of superficial skin wounds and to evaluate its accuracy compared to standard manual measurement methods.    Material and methods.  The hardware of the platform was developed using an Arduino Uno microcontroller, 28BYJ-48 stepper motors, and a Sharp 2Y0A21 infrared distance sensor. The software controls the scanning process and calculates wound parameters: volume (using Simpson’s method), area (using the Gaussian formula), and maximum depth. For validation, six model wounds with varying geometries were created on porcine skin samples. Their parameters were independently measured by the automated system and manual methods: volume by filling with saline solution, depth with a caliper, and area by planimetry in ImageJ software.    Results.  Statistical analysis demonstrated high agreement between automated and manual measurements. The highest accuracy was achieved in volume assessment: the mean difference between methods (0.37 cm 3 , p = 0.156), Spearman’s (ρ = 1.00) and Lin’s concordance correlation coefficient (> 0.98) indicated nearly perfect concordance. Depth measurements also showed high reliability, despite a minor systematic underestimation by the automated system averaging 0.99 mm (p = 0.094). A strong monotonic relationship was confirmed (ρ = 0.94). For wound area, a statistically significant systematic bias was identified: the platform consistently overestimated values by an average of 1.98 cm 2 (p = 0.031), likely due to hardware limitations and insufficient system resolution.    Conclusions.  The developed platform proved effective for automated, non-contact measurement of wound volume and depth, combining component affordability and automation. The main limitation is a systematic error in area measurement. Future work involves hardware modernization (installing a high-precision laser sensor) and further validation on an expanded sample of various wound types for implementation in clinical and research practice.