Adrian B. Chlopowiec
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Adrian B. Chlopowiec has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Martin Tabakov
Wroclaw University of Science and Technology
4 shared publications - Adam R. Chlopowiec
Department of Artificial Intelligence, Wroclaw University of Science and Technology, Wybrzeże Wyspianskiego 27, 50-370 Wroclaw, Poland
3 shared publications - Adam Dlubak
Department of Computational Intelligence, Wroclaw University of Science and Technology, 50-370 Wroclaw, Poland
1 shared publication - Konrad Karanowski
Department of Artificial Intelligence, Wroclaw University of Science and Technology, Wrocław, Poland
1 shared publication - Konrad Karanowski
Department of Artificial Intelligence, Wroclaw University of Science and Technology, Wybrzeże Wyspianskiego 27, 50-370 Wroclaw, Poland
1 shared publication - Mateusz Grzesiuk
Department of Artificial Intelligence, Wroclaw University of Science and Technology, Wybrzeże Wyspianskiego 27, 50-370 Wroclaw, Poland
1 shared publication
Latest Publications
- Open Access
- 29 May 2023
- Published by MDPI AG in Diagnostics
- Vol. 13 (11) , 1904
Counteracting Data Bias and Class Imbalance—Towards a Useful and Reliable Retinal Disease Recognition System
Multiple studies presented satisfactory performances for the treatment of various ocular diseases. To date, there has been no study that describes a multiclass model, medically accurate, and trained on large diverse dataset. No study has addressed a class imbalance problem in one giant dataset originating from multiple large diverse eye fundus image collections. To ensure a real-life clinical environment and mitigate the problem of biased medical image data, 22 publicly available datasets were merged. To secure medical validity only Diabetic Retinopathy (DR), Age-Related Macular Degeneration (AMD) and Glaucoma (GL) were included. The state-of-the-art models ConvNext, RegNet and ResNet were utilized. In the resulting dataset, there were 86,415 normal, 3787 GL, 632 AMD and 34,379 DR fundus images. ConvNextTiny achieved the best results in terms of recognizing most of the examined eye diseases with the most metrics. The overall accuracy was 80.46 ± 1.48. Specific accuracy values were: 80.01 ± 1.10 for normal eye fundus, 97.20 ± 0.66 for GL, 98.14 ± 0.31 for AMD, 80.66 ± 1.27 for DR. A suitable screening model for the most prevalent retinal diseases in ageing societies was designed. The model was developed on a diverse, combined large dataset which made the obtained results less biased and more generalizable.
- 29 May 2023
- Published by MDPI AG in Diagnostics
- Vol. 13 (11) , 1904
Open AccessCounteracting Data Bias and Class Imbalance—Towards a Useful and Reliable Retinal Disease Recognition System
Multiple studies presented satisfactory performances for the treatment of various ocular diseases. To date, there has been no study that describes a multiclass model, medically accurate, and trained on large diverse dataset. No study has addressed a class imbalance problem in one giant dataset originating from multiple large diverse eye fundus image collections. To ensure a real-life clinical environment and mitigate the problem of biased medical image data, 22 publicly available datasets were merged. To secure medical validity only Diabetic Retinopathy (DR), Age-Related Macular Degeneration (AMD) and Glaucoma (GL) were included. The state-of-the-art models ConvNext, RegNet and ResNet were utilized. In the resulting dataset, there were 86,415 normal, 3787 GL, 632 AMD and 34,379 DR fundus images. ConvNextTiny achieved the best results in terms of recognizing most of the examined eye diseases with the most metrics. The overall accuracy was 80.46 ± 1.48. Specific accuracy values were: 80.01 ± 1.10 for normal eye fundus, 97.20 ± 0.66 for GL, 98.14 ± 0.31 for AMD, 80.66 ± 1.27 for DR. A suitable screening model for the most prevalent retinal diseases in ageing societies was designed. The model was developed on a diverse, combined large dataset which made the obtained results less biased and more generalizable.
- Open Access
- 23 April 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (9) , 5279
A Novel Classification Method Using the Takagi–Sugeno Model and a Type-2 Fuzzy Rule Induction Approach
The main purpose of this research was to introduce a classification method, which combines a rule induction procedure with the Takagi–Sugeno inference model. This proposal is a continuation of our previous research, in which a classification process based on interval type-2 fuzzy rule induction was introduced. The research goal was to verify if the Mamdani fuzzy inference used in our previous research could be replaced with the first-order Takagi–Sugeno inference system. In the both cases to induce fuzzy rules, a new concept of a fuzzy information system was defined in order to deal with interval type-2 fuzzy sets. Additionally, the introduced rule induction assumes an optimization procedure concerning the footprint of uncertainty of the considered type-2 fuzzy sets. A key point in the concept proposed is the generalization of the fuzzy information systems’ attribute information to handle uncertainty, which occurs in real data. For experimental purposes, the classification method was tested on different classification benchmark data and very promising results were achieved. For the data sets: Breast Cancer Data, Breast Cancer Wisconsin, Data Banknote Authentication, HTRU 2 and Ionosphere, the following F-scores were achieved, respectively: 97.6%, 96%, 100%, 87.8%, and 89.4%. The results proved the possibility of applying the Takagi–Sugeno model in the classification concept. The model parameters were optimized using an evolutionary strategy.
- 23 April 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (9) , 5279
Open AccessA Novel Classification Method Using the Takagi–Sugeno Model and a Type-2 Fuzzy Rule Induction Approach
The main purpose of this research was to introduce a classification method, which combines a rule induction procedure with the Takagi–Sugeno inference model. This proposal is a continuation of our previous research, in which a classification process based on interval type-2 fuzzy rule induction was introduced. The research goal was to verify if the Mamdani fuzzy inference used in our previous research could be replaced with the first-order Takagi–Sugeno inference system. In the both cases to induce fuzzy rules, a new concept of a fuzzy information system was defined in order to deal with interval type-2 fuzzy sets. Additionally, the introduced rule induction assumes an optimization procedure concerning the footprint of uncertainty of the considered type-2 fuzzy sets. A key point in the concept proposed is the generalization of the fuzzy information systems’ attribute information to handle uncertainty, which occurs in real data. For experimental purposes, the classification method was tested on different classification benchmark data and very promising results were achieved. For the data sets: Breast Cancer Data, Breast Cancer Wisconsin, Data Banknote Authentication, HTRU 2 and Ionosphere, the following F-scores were achieved, respectively: 97.6%, 96%, 100%, 87.8%, and 89.4%. The results proved the possibility of applying the Takagi–Sugeno model in the classification concept. The model parameters were optimized using an evolutionary strategy.
- 21 September 2022
- Published by Springer Nature
- p. 95-105
Data Augmentation for Morphological Analysis of Histopathological Images Using Deep Learning
- 21 September 2022
- Published by Springer Nature
- p. 95-105
Data Augmentation for Morphological Analysis of Histopathological Images Using Deep Learning
- Open Access
- 13 April 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (8) , 3484
Classification with Fuzzification Optimization Combining Fuzzy Information Systems and Type-2 Fuzzy Inference
In this research, we introduce a classification procedure based on rule induction and fuzzy reasoning. The classifier generalizes attribute information to handle uncertainty, which often occurs in real data. To induce fuzzy rules, we define the corresponding fuzzy information system. A transformation of the derived rules into interval type-2 fuzzy rules is provided as well. The fuzzification applied is optimized with respect to the footprint of uncertainty of the corresponding type-2 fuzzy sets. The classification process is related to a Mamdani type fuzzy inference. The method proposed was evaluated by the F-score measure on benchmark data.
- 13 April 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (8) , 3484
Open AccessClassification with Fuzzification Optimization Combining Fuzzy Information Systems and Type-2 Fuzzy Inference
In this research, we introduce a classification procedure based on rule induction and fuzzy reasoning. The classifier generalizes attribute information to handle uncertainty, which often occurs in real data. To induce fuzzy rules, we define the corresponding fuzzy information system. A transformation of the derived rules into interval type-2 fuzzy rules is provided as well. The fuzzification applied is optimized with respect to the footprint of uncertainty of the corresponding type-2 fuzzy sets. The classification process is related to a Mamdani type fuzzy inference. The method proposed was evaluated by the F-score measure on benchmark data.