Konstantin Kiselev - Academia.edu (original) (raw)

Uploads

Papers by Konstantin Kiselev

Research paper thumbnail of Stack-U-Net: Refinement Network for Image Segmentation on the Example of Optic Disc and Cup

In this work, we propose a special cascade network for image segmentation, which is based on the ... more In this work, we propose a special cascade network for image segmentation, which is based on the U-Net networks as building blocks and the idea of the iterative refinement. The model was mainly applied to achieve higher recognition quality for the task of finding borders of the optic disc and cup, which are relevant to the presence of glaucoma. Compared to a single U-Net and the state-of-the-art methods for the investigated tasks, very high segmentation quality has been achieved without a need for increasing the volume of datasets. Our experiments include comparison with the best-known methods on publicly available databases DRIONS-DB, RIM-ONE v.3, DRISHTI-GS, and evaluation on a private data set collected in collaboration with University of California San Francisco Medical School. The analysis of the architecture details is presented, and it is argued that the model can be employed for a broad scope of image segmentation problems of similar nature.

Research paper thumbnail of Stack-U-Net: refinement network for improved optic disc and cup image segmentation

Medical Imaging 2019: Image Processing, 2019

In this work, we propose a special cascade network for image segmentation, which is based on the ... more In this work, we propose a special cascade network for image segmentation, which is based on the U-Net networks as building blocks and the idea of the iterative refinement. The model was mainly applied to achieve higher recognition quality for the task of finding borders of the optic disc and cup, which are relevant to the presence of glaucoma. Compared to a single U-Net and the state-of-the-art methods for the investigated tasks, the presented method outperforms others by multiple benchmarks without a need for increasing the volume of datasets. Our experiments include comparison with the best-known methods on publicly available databases DRIONS-DB, RIM-ONE v.3, DRISHTI-GS, and evaluation on a private data set collected in collaboration with University of California San Francisco Medical School. The analysis of the architecture details is presented. It is argued that the model can be employed for a broad scope of image segmentation problems of similar nature.

Research paper thumbnail of PhotoAgeClock: deep learning algorithms for development of non-invasive visual biomarkers of aging

Aging, 2018

Aging biomarkers are the qualitative and quantitative indicators of the aging processes of the hu... more Aging biomarkers are the qualitative and quantitative indicators of the aging processes of the human body. Estimation of biological age is important for assessing the physiological state of an organism. The advent of machine learning lead to the development of the many age predictors commonly referred to as the "aging clocks" varying in biological relevance, ease of use, cost, actionability, interpretability, and applications. Here we present and investigate a novel non-invasive class of visual photographic biomarkers of aging. We developed a simple and accurate predictor of chronological age using just the anonymized images of eye corners called the PhotoAgeClock. Deep neural networks were trained on 8414 anonymized high-resolution images of eye corners labeled with the correct chronological age. For people within the age range of 20 to 80 in a specific population, the model was able to achieve a mean absolute error of 2.3 years and 95% Pearson and Spearman correlation.

Research paper thumbnail of 14086 A new multimodal age prediction image analysis method from hands images of different age groups by neural network model

Journal of the American Academy of Dermatology, 2020

Four types of adverse events resembling immunological reactions have been reported in patients un... more Four types of adverse events resembling immunological reactions have been reported in patients undergoing chronic maintenance hemodialysis: anaphylaxis [1], hypereosinophilia [1-5], asthmatic attacks [4-6], and pulmonary leukostasis with compromised pulmonary function [7-10]. We evaluated three successive chronic dialysis patients who had developed acute hypersensitivity reactions manifested by increased airway obstruction in response to hemodialysis. Each patient also had a non-acute reaction to dialysis, hypereosinophilia, and one patient had urticaria. All dialyzers were ethylene oxide sterilized and none were reused. In these three patients the adverse reactions abated after changing from dialyzers containing Cuprophan (ENKA Ag, Wuppertal, Germany) to dialyzers not containing Cuprophan. In each patient a second exposure to Cuprophan caused recurrence of one or more of the adverse reactions. These case studies and our re-examination of previous observations [1-61 suggest that in some patients both acute and chronic adverse reactions may be related to the use of Cuprophan. Methods and Results. Patient 1. A 50-year-old black female with renal failure secondary to hypertension began chronic hemodialysis in December 1975. It is uncertain which dialyzers were used during the first 10 months of dialysis, but during her ninth and tenth months of dialysis differential counts showed 18 and 20% eosinophils (Fig. 1). In October 1976, a one antigenmatched cadaver kidney transplant was attempted but was rejected and, 3 weeks later, removed. Hemodialysis was continued using coil dialyzers containing Cuprophan (ENKA Ag)

Research paper thumbnail of Stack-U-Net: Refinement Network for Image Segmentation on the Example of Optic Disc and Cup

In this work, we propose a special cascade network for image segmentation, which is based on the ... more In this work, we propose a special cascade network for image segmentation, which is based on the U-Net networks as building blocks and the idea of the iterative refinement. The model was mainly applied to achieve higher recognition quality for the task of finding borders of the optic disc and cup, which are relevant to the presence of glaucoma. Compared to a single U-Net and the state-of-the-art methods for the investigated tasks, very high segmentation quality has been achieved without a need for increasing the volume of datasets. Our experiments include comparison with the best-known methods on publicly available databases DRIONS-DB, RIM-ONE v.3, DRISHTI-GS, and evaluation on a private data set collected in collaboration with University of California San Francisco Medical School. The analysis of the architecture details is presented, and it is argued that the model can be employed for a broad scope of image segmentation problems of similar nature.

Research paper thumbnail of Stack-U-Net: refinement network for improved optic disc and cup image segmentation

Medical Imaging 2019: Image Processing, 2019

In this work, we propose a special cascade network for image segmentation, which is based on the ... more In this work, we propose a special cascade network for image segmentation, which is based on the U-Net networks as building blocks and the idea of the iterative refinement. The model was mainly applied to achieve higher recognition quality for the task of finding borders of the optic disc and cup, which are relevant to the presence of glaucoma. Compared to a single U-Net and the state-of-the-art methods for the investigated tasks, the presented method outperforms others by multiple benchmarks without a need for increasing the volume of datasets. Our experiments include comparison with the best-known methods on publicly available databases DRIONS-DB, RIM-ONE v.3, DRISHTI-GS, and evaluation on a private data set collected in collaboration with University of California San Francisco Medical School. The analysis of the architecture details is presented. It is argued that the model can be employed for a broad scope of image segmentation problems of similar nature.

Research paper thumbnail of PhotoAgeClock: deep learning algorithms for development of non-invasive visual biomarkers of aging

Aging, 2018

Aging biomarkers are the qualitative and quantitative indicators of the aging processes of the hu... more Aging biomarkers are the qualitative and quantitative indicators of the aging processes of the human body. Estimation of biological age is important for assessing the physiological state of an organism. The advent of machine learning lead to the development of the many age predictors commonly referred to as the "aging clocks" varying in biological relevance, ease of use, cost, actionability, interpretability, and applications. Here we present and investigate a novel non-invasive class of visual photographic biomarkers of aging. We developed a simple and accurate predictor of chronological age using just the anonymized images of eye corners called the PhotoAgeClock. Deep neural networks were trained on 8414 anonymized high-resolution images of eye corners labeled with the correct chronological age. For people within the age range of 20 to 80 in a specific population, the model was able to achieve a mean absolute error of 2.3 years and 95% Pearson and Spearman correlation.

Research paper thumbnail of 14086 A new multimodal age prediction image analysis method from hands images of different age groups by neural network model

Journal of the American Academy of Dermatology, 2020

Four types of adverse events resembling immunological reactions have been reported in patients un... more Four types of adverse events resembling immunological reactions have been reported in patients undergoing chronic maintenance hemodialysis: anaphylaxis [1], hypereosinophilia [1-5], asthmatic attacks [4-6], and pulmonary leukostasis with compromised pulmonary function [7-10]. We evaluated three successive chronic dialysis patients who had developed acute hypersensitivity reactions manifested by increased airway obstruction in response to hemodialysis. Each patient also had a non-acute reaction to dialysis, hypereosinophilia, and one patient had urticaria. All dialyzers were ethylene oxide sterilized and none were reused. In these three patients the adverse reactions abated after changing from dialyzers containing Cuprophan (ENKA Ag, Wuppertal, Germany) to dialyzers not containing Cuprophan. In each patient a second exposure to Cuprophan caused recurrence of one or more of the adverse reactions. These case studies and our re-examination of previous observations [1-61 suggest that in some patients both acute and chronic adverse reactions may be related to the use of Cuprophan. Methods and Results. Patient 1. A 50-year-old black female with renal failure secondary to hypertension began chronic hemodialysis in December 1975. It is uncertain which dialyzers were used during the first 10 months of dialysis, but during her ninth and tenth months of dialysis differential counts showed 18 and 20% eosinophils (Fig. 1). In October 1976, a one antigenmatched cadaver kidney transplant was attempted but was rejected and, 3 weeks later, removed. Hemodialysis was continued using coil dialyzers containing Cuprophan (ENKA Ag)