Nikolay Evgenievich Staroverov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Nikolay Evgenievich Staroverov has a total of 69 co-authors. Here are the top co-authors by number of shared publications:
- Kirill Tkachenko
Komarov Botanical Institute of RAS (BIN) , Saint Petersburg , Russia
7 shared publications - Nikolay Nefediev
Department of Bioinformatics and Mathematical Biology, Alferov University, 194021 St. Petersburg, Russia, Russia
5 shared publications - Nikolay Nikolaevich Potrakhov
Saint Petersburg Electrotechnical University
5 shared publications - Artem Gryaznov
Saint Petersburg Electrotechnical University st. Professora Popova 5, 197376 St. Petersburg, Russian Federation
4 shared publications - Roman V. Davydov
Higher School of Cyber-Physical Systems & Control, Peter the Great St. Petersburg Polytechnic University, 195251 St. Petersburg, Russia
4 shared publications A Yu Gryaznov
Department of electronic instruments and devices, Saint Petersburg Electrotechnical University "LETI", 197376, Saint Petersburg, Russia
3 shared publications
Latest Publications
- Open Access
- 4 November 2026
- Published by MDPI AG in Preprints.org
Phase-Mixed Versus Phase-Specific Training for Pancreatic Lesion Segmentation on Contrast-Enhanced CT: A Non-Inferiority Study on PanTS
A model that segments pancreatic lesions from either arterial or portal venous CT could simplify deployment, provided that its performance is not inferior to that of phase-specific models. We evaluated nnU-Net ResEnc-M models trained on arterial scans (A), portal venous scans (P), or the combined dataset (M-full). Quality control, reconstruction of patient groups, and automatic phase assignment yielded 3,085 development scans and 438 held-out PanTS test scans. Five-fold cross-validation was repeated with three training seeds for A, P, and M-full; five reduced, phase-balanced training subsets were evaluated using seed 0. Non-inferiority required the lower confidence bound for the paired difference in lesion Dice to exceed −0.03. M-full met this criterion in both phases during cross-validation with all three seeds. On 85 lesion-positive portal test scans, its Dice advantage over P was 0.0387–0.0473, with all corresponding 95% confidence intervals above zero. M-full had fewer complete misses (17–19 versus 21–23 of 85), but more false-positive scans (34–39 versus 23–27 of 243 lesion-negative scans). Reducing the portal training subset lowered portal cross-validation performance in all five balanced variants. The arterial test contained only four positive scans and was therefore descriptive. Evaluation on 281 annotated Task07 scans was also descriptive because of known and unresolved dataset overlap. Mixed-phase training met the specified non-inferiority criterion while requiring only one CT volume at inference, but small-lesion misses and the false-positive trade-off remained.
- 4 November 2026
- Published by MDPI AG in Preprints.org
Open AccessPhase-Mixed Versus Phase-Specific Training for Pancreatic Lesion Segmentation on Contrast-Enhanced CT: A Non-Inferiority Study on PanTS
A model that segments pancreatic lesions from either arterial or portal venous CT could simplify deployment, provided that its performance is not inferior to that of phase-specific models. We evaluated nnU-Net ResEnc-M models trained on arterial scans (A), portal venous scans (P), or the combined dataset (M-full). Quality control, reconstruction of patient groups, and automatic phase assignment yielded 3,085 development scans and 438 held-out PanTS test scans. Five-fold cross-validation was repeated with three training seeds for A, P, and M-full; five reduced, phase-balanced training subsets were evaluated using seed 0. Non-inferiority required the lower confidence bound for the paired difference in lesion Dice to exceed −0.03. M-full met this criterion in both phases during cross-validation with all three seeds. On 85 lesion-positive portal test scans, its Dice advantage over P was 0.0387–0.0473, with all corresponding 95% confidence intervals above zero. M-full had fewer complete misses (17–19 versus 21–23 of 85), but more false-positive scans (34–39 versus 23–27 of 243 lesion-negative scans). Reducing the portal training subset lowered portal cross-validation performance in all five balanced variants. The arterial test contained only four positive scans and was therefore descriptive. Evaluation on 281 annotated Task07 scans was also descriptive because of known and unresolved dataset overlap. Mixed-phase training met the specified non-inferiority criterion while requiring only one CT volume at inference, but small-lesion misses and the false-positive trade-off remained.
- 14 September 2026
- Published by Pleiades Publishing Ltd in Bulletin of the Russian Academy of Sciences: Physics
- Vol. 90 (9) , 1531-1538
Application of Convolutional Neural Networks to Improve Tumor Segmentation in Prostate MRI
The study aims to obtain segmentations (masks) of areas with prostate cancer in MRI studies of the pelvic organs in men. Various neural network architectures for 2D and 3D segmentation in MRI series were examined, as well as methods for improving results of prostate presegmentation using a classification neural network. Methods for improving prostate segmentation were also considered.
- 14 September 2026
- Published by Pleiades Publishing Ltd in Bulletin of the Russian Academy of Sciences: Physics
- Vol. 90 (9) , 1531-1538
Application of Convolutional Neural Networks to Improve Tumor Segmentation in Prostate MRI
The study aims to obtain segmentations (masks) of areas with prostate cancer in MRI studies of the pelvic organs in men. Various neural network architectures for 2D and 3D segmentation in MRI series were examined, as well as methods for improving results of prostate presegmentation using a classification neural network. Methods for improving prostate segmentation were also considered.
- 30 June 2026
- Published by FSBSI FRC N.I. Vavilov All-Russian Institute of Plant Genetic Resources in Proceedings on Applied Botany, Genetics and Breeding
- Vol. 187 (2) , 105-113
Latent period characteristics of Iris pallasii Fisch. ex Trevir. (Iridaceae)
Background . Species of the genus Iris L. attract the attention of researchers as a source of bioactive compounds promising for the development of medicinal drugs. There are no data on the characteristics of the latent period for Iris pallasii Fisch. ex Trevir. in the flora of Kazakhstan. Materials and methods. Seeds were collected from 2015 through 2022 in Almaty Region and from plants grown in the medicinal plant collection plot of the Botanical Garden of the Institute of Botany and Phytointroduction. A soft-beam mobile X-ray diagnostic unit was used to assess the quality of I. pallasii seeds. A 0.3% hydrogen peroxide solution, a 0.05% epibrassinolide solution, and Biostim were applied as standard growth stimulants. Control plants were germinated in water only. Results and discussion. The 1000-seed weight varied slightly within the species, regardless of the origin of seeds: natural I. pallasii seeds weighed from 17.2 to 19.1 g, while introduced plants of this species produced slightly larger seeds, weighing from 18.1 to 19.7 g. Laboratory evaluation of I. pallasii seed germination showed that the use of various growth stimulants contributed to increased germination, which was evident during prolonged seed germination (up to 110 days). The maximum number of germinating seeds was recorded during the 30–40 (up to 80) days of the experiment. However, some seeds (in both the control and experimental plants) germinated only on days 100–110 of the experiment. Conclusion . I. pallasii produced seeds of varying quality in different years. Under cultivation conditions at the Botanical Garden, they typically produced a higher percentage of mature, viable seeds with high, long-lasting germination. Growth stimulants increased laboratory seed germination by 10–20%. The maximum number of germinating I. pallasii seeds under all experimental conditions, regardless of growing location, was observed between days 40 and 80.
- 30 June 2026
- Published by FSBSI FRC N.I. Vavilov All-Russian Institute of Plant Genetic Resources in Proceedings on Applied Botany, Genetics and Breeding
- Vol. 187 (2) , 105-113
Latent period characteristics of Iris pallasii Fisch. ex Trevir. (Iridaceae)
Background . Species of the genus Iris L. attract the attention of researchers as a source of bioactive compounds promising for the development of medicinal drugs. There are no data on the characteristics of the latent period for Iris pallasii Fisch. ex Trevir. in the flora of Kazakhstan. Materials and methods. Seeds were collected from 2015 through 2022 in Almaty Region and from plants grown in the medicinal plant collection plot of the Botanical Garden of the Institute of Botany and Phytointroduction. A soft-beam mobile X-ray diagnostic unit was used to assess the quality of I. pallasii seeds. A 0.3% hydrogen peroxide solution, a 0.05% epibrassinolide solution, and Biostim were applied as standard growth stimulants. Control plants were germinated in water only. Results and discussion. The 1000-seed weight varied slightly within the species, regardless of the origin of seeds: natural I. pallasii seeds weighed from 17.2 to 19.1 g, while introduced plants of this species produced slightly larger seeds, weighing from 18.1 to 19.7 g. Laboratory evaluation of I. pallasii seed germination showed that the use of various growth stimulants contributed to increased germination, which was evident during prolonged seed germination (up to 110 days). The maximum number of germinating seeds was recorded during the 30–40 (up to 80) days of the experiment. However, some seeds (in both the control and experimental plants) germinated only on days 100–110 of the experiment. Conclusion . I. pallasii produced seeds of varying quality in different years. Under cultivation conditions at the Botanical Garden, they typically produced a higher percentage of mature, viable seeds with high, long-lasting germination. Growth stimulants increased laboratory seed germination by 10–20%. The maximum number of germinating I. pallasii seeds under all experimental conditions, regardless of growing location, was observed between days 40 and 80.
- Open Access
- 18 June 2026
- Published by ECO-Vector LLC (Publications) in Journal of Clinical Practice
- Vol. 17 (2) , 55-68
CT-perfusion parameters and reperfusion therapy in acute ischemic stroke: 24-hour NIHSS dynamics in a retrospective cohort.
BACKGROUND: Neurological improvement after acute ischemic stroke depends on multiple clinical and imaging factors, including baseline stroke severity, infarct volume, and the success of reperfusion therapy. Despite technically successful recanalization, 30–50% of patients show no meaningful clinical improvement. AIM: To evaluate the impact of CT perfusion parameters, intravenous thrombolysis, and mechanical thrombectomy on 24‑hour neurological status dynamics measured by the NIHSS in patients with acute ischemic stroke due to large vessel occlusion. METHODS: This retrospective single‑center cohort included 52 patients with acute ischemic stroke caused by large vessel occlusion. CT perfusion data were analyzed using the ArchiMed PRO software. Ridge regression was applied to assess the effects of perfusion parameters, reperfusion therapies, and their characteristics on 24‑hour change in NIHSS. RESULTS: The model explained 44% of the variance in NIHSS change. Statistically significant predictors were baseline NIHSS score (p0.05), degree of recanalization on the TICI scale (p0.05; positive effect), and the performance of mechanical thrombectomy (p0.05; negative effect). Non‑significant trends were observed for infarct core volume (negative effect, p0.05), penumbra volume (positive effect, p0.05), and intravenous thrombolysis (positive effect, p0.05). CONCLUSION: Successful recanalization is the leading predictor of short‑term neurological improvement after acute ischemic stroke. CT perfusion parameters show trends consistent with previous literature; however, the limited statistical power of this cohort prevented these associations from reaching formal statistical significance.
- 18 June 2026
- Published by ECO-Vector LLC (Publications) in Journal of Clinical Practice
- Vol. 17 (2) , 55-68
Open AccessCT-perfusion parameters and reperfusion therapy in acute ischemic stroke: 24-hour NIHSS dynamics in a retrospective cohort.
BACKGROUND: Neurological improvement after acute ischemic stroke depends on multiple clinical and imaging factors, including baseline stroke severity, infarct volume, and the success of reperfusion therapy. Despite technically successful recanalization, 30–50% of patients show no meaningful clinical improvement. AIM: To evaluate the impact of CT perfusion parameters, intravenous thrombolysis, and mechanical thrombectomy on 24‑hour neurological status dynamics measured by the NIHSS in patients with acute ischemic stroke due to large vessel occlusion. METHODS: This retrospective single‑center cohort included 52 patients with acute ischemic stroke caused by large vessel occlusion. CT perfusion data were analyzed using the ArchiMed PRO software. Ridge regression was applied to assess the effects of perfusion parameters, reperfusion therapies, and their characteristics on 24‑hour change in NIHSS. RESULTS: The model explained 44% of the variance in NIHSS change. Statistically significant predictors were baseline NIHSS score (p0.05), degree of recanalization on the TICI scale (p0.05; positive effect), and the performance of mechanical thrombectomy (p0.05; negative effect). Non‑significant trends were observed for infarct core volume (negative effect, p0.05), penumbra volume (positive effect, p0.05), and intravenous thrombolysis (positive effect, p0.05). CONCLUSION: Successful recanalization is the leading predictor of short‑term neurological improvement after acute ischemic stroke. CT perfusion parameters show trends consistent with previous literature; however, the limited statistical power of this cohort prevented these associations from reaching formal statistical significance.
- Open Access
- 9 April 2026
- Published by ECO-Vector LLC (Publications) in Journal of Clinical Practice
- Vol. 17 (1) , 50-61
Validation of a Russian automated CT perfusion processing system for ischemic stroke
BACKGROUND: Automated computed tomography (CT) perfusion analysis systems are critical for patient selection for reperfusion therapy in acute ischemic stroke. Systems such as RAPID (iSchemaView/RapidAI, USA) and Olea Sphere (Olea Medical, France) are widely used in clinical practice; however, a validated domestic system for automated CT perfusion processing is lacking in the Russian Federation. Among similar systems that have a registration certificate in the Russian Federation, Vitrea (Canon Medical Systems, USA, Japan) is available in clinical practice. AIM: To validate the Russian system ArchiMed PRO Chronos by comparing it with the Vitrea (Canon Medical Systems, USA, Japan) software for quantitative assessment of ischemic core and critical hypoperfusion area. METHODS: A retrospective study included 52 patients with acute ischemic stroke caused by large vessel occlusion and symptom onset within 24 hours. Correlation between systems was assessed for ischemic core volume and critical hypoperfusion area determination. RESULTS: Strong correlation was found between systems for ischemic core volume (r=0.95) and critical hypoperfusion area (r=0.85). After excluding studies with inadequate segmentation, correlation improved to r=0.98 and r=0.93, respectively. Adequate segmentation was achieved in 84.6% of cases for both systems. No statistically significant differences between systems were detected (p 0.05). CONCLUSION: The ArchiMed PRO Chronos system demonstrated excellent agreement with Vitrea and can be used for quantitative CT perfusion analysis and patient selection for reperfusion therapy.
- 9 April 2026
- Published by ECO-Vector LLC (Publications) in Journal of Clinical Practice
- Vol. 17 (1) , 50-61
Open AccessValidation of a Russian automated CT perfusion processing system for ischemic stroke
BACKGROUND: Automated computed tomography (CT) perfusion analysis systems are critical for patient selection for reperfusion therapy in acute ischemic stroke. Systems such as RAPID (iSchemaView/RapidAI, USA) and Olea Sphere (Olea Medical, France) are widely used in clinical practice; however, a validated domestic system for automated CT perfusion processing is lacking in the Russian Federation. Among similar systems that have a registration certificate in the Russian Federation, Vitrea (Canon Medical Systems, USA, Japan) is available in clinical practice. AIM: To validate the Russian system ArchiMed PRO Chronos by comparing it with the Vitrea (Canon Medical Systems, USA, Japan) software for quantitative assessment of ischemic core and critical hypoperfusion area. METHODS: A retrospective study included 52 patients with acute ischemic stroke caused by large vessel occlusion and symptom onset within 24 hours. Correlation between systems was assessed for ischemic core volume and critical hypoperfusion area determination. RESULTS: Strong correlation was found between systems for ischemic core volume (r=0.95) and critical hypoperfusion area (r=0.85). After excluding studies with inadequate segmentation, correlation improved to r=0.98 and r=0.93, respectively. Adequate segmentation was achieved in 84.6% of cases for both systems. No statistically significant differences between systems were detected (p 0.05). CONCLUSION: The ArchiMed PRO Chronos system demonstrated excellent agreement with Vitrea and can be used for quantitative CT perfusion analysis and patient selection for reperfusion therapy.
- Top Cited PapersOpen Access
- 19 January 2026
- Published by MDPI AG in Algorithms
- Vol. 19 (1) , 85
Improving Prostate Cancer Segmentation on T2-Weighted MRI Using Prostate Detection and Cascaded Networks
Prostate cancer is one of the most lethal cancers in the male population, and accurate localization of intraprostatic lesions on MRI remains challenging. In this study, we investigated methods for improving prostate cancer segmentation on T2-weighted pelvic MRI using cascaded neural networks. We used an anonymized dataset of 400 multiparametric MRI scans from two centers, in which experienced radiologists had delineated the prostate and clinically significant cancer on the T2 series. Our baseline approach applies 2D and 3D segmentation networks (UNETR, UNET++, Swin-UNETR, SegResNetDS, and SegResNetVAE) directly to full MRI volumes. We then introduce additional stages that filter slices using DenseNet-201 classifiers (cancer/no-cancer and prostate/no-prostate) and localize the prostate via a YOLO-based detector to crop the 3D region of interest before segmentation. Using Swin-UNETR as the backbone, the prostate segmentation Dice score increased from 71.37% for direct 3D segmentation to 76.09% when using prostate detection and cropped 3D inputs. For cancer segmentation, the final cascaded pipeline—prostate detection, 3D prostate segmentation, and 3D cancer segmentation within the prostate—improved the Dice score from 55.03% for direct 3D segmentation to 67.11%, with an ROC AUC of 0.89 on the test set. These results suggest that cascaded detection- and segmentation-based preprocessing of the prostate region can substantially improve automatic prostate cancer segmentation on MRI while remaining compatible with standard segmentation architectures.
- 19 January 2026
- Published by MDPI AG in Algorithms
- Vol. 19 (1) , 85
Top Cited PapersOpen AccessImproving Prostate Cancer Segmentation on T2-Weighted MRI Using Prostate Detection and Cascaded Networks
Prostate cancer is one of the most lethal cancers in the male population, and accurate localization of intraprostatic lesions on MRI remains challenging. In this study, we investigated methods for improving prostate cancer segmentation on T2-weighted pelvic MRI using cascaded neural networks. We used an anonymized dataset of 400 multiparametric MRI scans from two centers, in which experienced radiologists had delineated the prostate and clinically significant cancer on the T2 series. Our baseline approach applies 2D and 3D segmentation networks (UNETR, UNET++, Swin-UNETR, SegResNetDS, and SegResNetVAE) directly to full MRI volumes. We then introduce additional stages that filter slices using DenseNet-201 classifiers (cancer/no-cancer and prostate/no-prostate) and localize the prostate via a YOLO-based detector to crop the 3D region of interest before segmentation. Using Swin-UNETR as the backbone, the prostate segmentation Dice score increased from 71.37% for direct 3D segmentation to 76.09% when using prostate detection and cropped 3D inputs. For cancer segmentation, the final cascaded pipeline—prostate detection, 3D prostate segmentation, and 3D cancer segmentation within the prostate—improved the Dice score from 55.03% for direct 3D segmentation to 67.11%, with an ROC AUC of 0.89 on the test set. These results suggest that cascaded detection- and segmentation-based preprocessing of the prostate region can substantially improve automatic prostate cancer segmentation on MRI while remaining compatible with standard segmentation architectures.
- Open Access
- 22 December 2025
- Published by MDPI AG in Preprints.org
Improving Prostate Cancer Segmentation on T2-Weighted MRI Using Prostate Detection and Cascaded Networks
Prostate cancer is one of the most lethal cancers in the male population, and accurate localization of intraprostatic lesions on MRI remains challenging. In this work, we investigate methods for improving prostate cancer segmentation on T2weighted pelvic MRI using cascaded neural networks. We use anonymized dataset of 400 multiparametric MRI studies from two centers, in which experienced radiologists delineated the prostate and clinically significant cancer on the T2 series. Our baseline approach applies 2D and 3D segmentation networks (UNETR, UNET++, SwinUNETR, SegResNetDS, SegResNetVAE) directly to full MRI volumes. We then introduce additional stages that filter slices using DenseNet201 classifiers (cancer/nocancer and prostate/noprostate) and localize the prostate with a YOLObased detector to crop a 3D region of interest before segmentation. Using SwinUNETR as the backbone, the prostate segmentation Dice score increased from 71.37% for direct 3D segmentation to 76.09% when using prostate detection and cropped 3D inputs. For cancer segmentation, the final cascaded pipeline – prostate detection, 3D prostate segmentation, and 3D cancer segmentation within the prostate – improved Dice score from 55.03% for direct 3D segmentation to 67.11%, with a ROC AUC of 0.89 on the test set. These results suggest that cascaded detection and segmentationbased preprocessing of the prostate region can substantially improve automatic prostate cancer segmentation on MRI while remaining compatible with standard segmentation architectures.
- 22 December 2025
- Published by MDPI AG in Preprints.org
Open AccessImproving Prostate Cancer Segmentation on T2-Weighted MRI Using Prostate Detection and Cascaded Networks
Prostate cancer is one of the most lethal cancers in the male population, and accurate localization of intraprostatic lesions on MRI remains challenging. In this work, we investigate methods for improving prostate cancer segmentation on T2weighted pelvic MRI using cascaded neural networks. We use anonymized dataset of 400 multiparametric MRI studies from two centers, in which experienced radiologists delineated the prostate and clinically significant cancer on the T2 series. Our baseline approach applies 2D and 3D segmentation networks (UNETR, UNET++, SwinUNETR, SegResNetDS, SegResNetVAE) directly to full MRI volumes. We then introduce additional stages that filter slices using DenseNet201 classifiers (cancer/nocancer and prostate/noprostate) and localize the prostate with a YOLObased detector to crop a 3D region of interest before segmentation. Using SwinUNETR as the backbone, the prostate segmentation Dice score increased from 71.37% for direct 3D segmentation to 76.09% when using prostate detection and cropped 3D inputs. For cancer segmentation, the final cascaded pipeline – prostate detection, 3D prostate segmentation, and 3D cancer segmentation within the prostate – improved Dice score from 55.03% for direct 3D segmentation to 67.11%, with a ROC AUC of 0.89 on the test set. These results suggest that cascaded detection and segmentationbased preprocessing of the prostate region can substantially improve automatic prostate cancer segmentation on MRI while remaining compatible with standard segmentation architectures.
- Open Access
- 7 August 2024
- Published by Vidar, Ltd. in Medical Visualization
- Vol. 28 (3) , 12-21
Deep Machine Learning for Automatic Segmentation of the Pancreatic Parenchyma and its hypo- and hypervascular lesions on CT Images
Objective of the study. To develop and evaluate the effectiveness of a technology for segmenting the pancreatic parenchyma and its hyper- and hypovascular lesions on abdominal computed tomography (CT) scans using deep machine learning. Materials and methods. CT scans from the database of the A.V. Vishnevsky National Medical Research Center of Surgery were used for training and testing the algorithms – a total number of approximately 150 studies (arterial and venous phases). A test dataset of 46 anonymized CT scans (arterial and venous phases) was prepared for validation of the obtained algorithms, independently assessed by expert physicians. The primary segmentation neural network used is nn-UNet (M. Antonelli et al., 2022). Results . The average accuracy of the test dataset for the model determining segmentation masks of the pancreas on CT images had an AUC of 0.8 for the venous phase and 0.85 for the arterial phase. The segmentation masks of pancreatic formations had an AUC of 0.6. Conclusion. Automated segmentation of the pancreatic parenchyma structure using deep machine learning technologies demonstrated high accuracy. However, the segmentation of hypo- and hypervascular pancreatic lesions requires improvement. The overlap of the masks showed a rather low result, but in all cases, the location of the pathological formation was correctly identified by the algorithm. Enhancing the training dataset and the algorithm used could increase the accuracy of the algorithm. No false negative results were obtained when detecting pancreatic formations; in all cases, the INS detected “suspicious” areas of the pancreatic parenchyma. This can help reduce the omission of pancreatic pathologies in CT scans, and their further assessment can be carried out by the radiologist himself.
- 7 August 2024
- Published by Vidar, Ltd. in Medical Visualization
- Vol. 28 (3) , 12-21
Open AccessDeep Machine Learning for Automatic Segmentation of the Pancreatic Parenchyma and its hypo- and hypervascular lesions on CT Images
Objective of the study. To develop and evaluate the effectiveness of a technology for segmenting the pancreatic parenchyma and its hyper- and hypovascular lesions on abdominal computed tomography (CT) scans using deep machine learning. Materials and methods. CT scans from the database of the A.V. Vishnevsky National Medical Research Center of Surgery were used for training and testing the algorithms – a total number of approximately 150 studies (arterial and venous phases). A test dataset of 46 anonymized CT scans (arterial and venous phases) was prepared for validation of the obtained algorithms, independently assessed by expert physicians. The primary segmentation neural network used is nn-UNet (M. Antonelli et al., 2022). Results . The average accuracy of the test dataset for the model determining segmentation masks of the pancreas on CT images had an AUC of 0.8 for the venous phase and 0.85 for the arterial phase. The segmentation masks of pancreatic formations had an AUC of 0.6. Conclusion. Automated segmentation of the pancreatic parenchyma structure using deep machine learning technologies demonstrated high accuracy. However, the segmentation of hypo- and hypervascular pancreatic lesions requires improvement. The overlap of the masks showed a rather low result, but in all cases, the location of the pathological formation was correctly identified by the algorithm. Enhancing the training dataset and the algorithm used could increase the accuracy of the algorithm. No false negative results were obtained when detecting pancreatic formations; in all cases, the INS detected “suspicious” areas of the pancreatic parenchyma. This can help reduce the omission of pancreatic pathologies in CT scans, and their further assessment can be carried out by the radiologist himself.
- Open Access
- 10 January 2024
- Published by FSBSI FRC N.I. Vavilov All-Russian Institute of Plant Genetic Resources in Proceedings on Applied Botany, Genetics and Breeding
- Vol. 184 (4) , 12-20
Fruit and seed plumpness, germination and viability of rare plants in Yakutia
Background. With the increasing anthropogenic impact on nature, conservation and renewal of rare and endangered plants in natural cenoses is an urgent problem. Quality assessment, identification of full-fledged plump seeds, and determination of their viability is the way to develop reproduction technologies and provide a possibility of repatriating these species to their natural habitats. Collecting and studying natural flora species in botanical gardens helps to save biological resources of useful species and recommend them for urban landscaping.Materials and methods. X-ray techniques were used to analyze seed quality and viability of rare species in the flora of Yakutia: Adonis sibirica Patrin, Phlojodicarpus sibiricus (Steph. ex Spreng.) Koso-Pol., Iris laevigata Fisch. et C.A. Mey., Iris sanguinea Donn, and Paeonia anomala L. Methods developed for the application of microfocus radiography to fruits and seeds were used in the X-ray analysis.Results. Seed quality was assessed according to guidelines. It was shown that fruits and seeds of the studied species developed under the conditions of Yakutia were full-fledged (plump), but due to the presence of a morphophysiological type of dormancy in them, they could germinate only after a cold or multistage stratification.Conclusion. Assessment of reproductive diaspore quality and viability in rare and endangered plant species of Yakutia will make it possible to develop technologies for increasing their seed germination rates and restoring the natural populations of these species.
- 10 January 2024
- Published by FSBSI FRC N.I. Vavilov All-Russian Institute of Plant Genetic Resources in Proceedings on Applied Botany, Genetics and Breeding
- Vol. 184 (4) , 12-20
Open AccessFruit and seed plumpness, germination and viability of rare plants in Yakutia
Background. With the increasing anthropogenic impact on nature, conservation and renewal of rare and endangered plants in natural cenoses is an urgent problem. Quality assessment, identification of full-fledged plump seeds, and determination of their viability is the way to develop reproduction technologies and provide a possibility of repatriating these species to their natural habitats. Collecting and studying natural flora species in botanical gardens helps to save biological resources of useful species and recommend them for urban landscaping.Materials and methods. X-ray techniques were used to analyze seed quality and viability of rare species in the flora of Yakutia: Adonis sibirica Patrin, Phlojodicarpus sibiricus (Steph. ex Spreng.) Koso-Pol., Iris laevigata Fisch. et C.A. Mey., Iris sanguinea Donn, and Paeonia anomala L. Methods developed for the application of microfocus radiography to fruits and seeds were used in the X-ray analysis.Results. Seed quality was assessed according to guidelines. It was shown that fruits and seeds of the studied species developed under the conditions of Yakutia were full-fledged (plump), but due to the presence of a morphophysiological type of dormancy in them, they could germinate only after a cold or multistage stratification.Conclusion. Assessment of reproductive diaspore quality and viability in rare and endangered plant species of Yakutia will make it possible to develop technologies for increasing their seed germination rates and restoring the natural populations of these species.
- Open Access
- 1 January 2024
- Published by Petrozavodsk State University in HORTUS BOTANICUS
- Vol. 19 (19) , 222-238
Picea schrenkiana Fisch. et C.A. Mey. in the Peter the Great Botanical Garden
- 1 January 2024
- Published by Petrozavodsk State University in HORTUS BOTANICUS
- Vol. 19 (19) , 222-238
Open AccessPicea schrenkiana Fisch. et C.A. Mey. in the Peter the Great Botanical Garden