Łukasz Czogalik

Mr. Łukasz Czogalik

Medical University of Silesia, Katowice, Śląskie, Poland
+1
orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
3 August 2023 in Diagnostics
Verified
Journal Article
Open access
PDF file
Jakub KufelJakub Kufel
Jakub Kufel
KBKatarzyna Bargieł-Łączek
Katarzyna Bargieł-Łączek
SKSzymon Kocot
Szymon Kocot
MKMaciej Koźlik
Maciej Koźlik
WBWiktoria Bartnikowska
Wiktoria Bartnikowska
KGKatarzyna Gruszczyńska
Katarzyna Gruszczyńska
Machine learning (ML), artificial neural networks (ANNs), and deep learning (DL) are all topics that fall under the heading of artificial intelligence (AI) and have gained popularity in recent years. ML involves the application of algorithms to automate decision-making processes using models that have not been manually programmed but have been trained on data. ANNs that are a part of ML aim to simulate the structure and function of the human brain. DL, on the other hand, uses multiple layers of interconnected neurons. This enables the processing and analysis of large and complex databases. In medicine, these techniques are being introduced to improve the speed and efficiency of disease diagnosis and treatment. Each of the AI techniques presented in the paper is supported with an example of a possible medical application. Given the rapid development of technology, the use of AI in medicine shows promising results in the context of patient care. It is particularly important to keep a close eye on this issue and conduct further research in order to fully explore the potential of ML, ANNs, and DL, and bring further applications into clinical use in the future.

22 September 2023 in Journal of Personalized Medicine
Verified
Journal Article
Open access
PDF file
Jakub KufelJakub Kufel
Jakub Kufel
MBMichał Bielówka
Michał Bielówka
MRMarcin Rojek
Marcin Rojek
AMAdam Mitręga
Adam Mitręga
PLPiotr Lewandowski
Piotr Lewandowski
ZNZbigniew Nawrat
Zbigniew Nawrat
In recent years, deep neural networks have enabled countless innovations in the field of image classification. Encouraged by success in this field, researchers worldwide have demonstrated how to use Convolutional Neural Network techniques in medical imaging problems. In this article, the results were obtained through the use of the EfficientNet in the task of classifying 14 different diseases based on chest X-ray images coming from the NIH (National Institutes of Health) ChestX-ray14 dataset. The approach addresses dataset imbalances by introducing a custom split to ensure fair representation. Binary cross entropy loss is utilized to handle the multi-label difficulty. The model architecture comprises an EfficientNet backbone for feature extraction, succeeded by sequential layers including GlobalAveragePooling, Dense, and BatchNormalization. The main contribution of this paper is a proposed solution that outperforms previous state-of-the-art deep learning models average area under the receiver operating characteristic curve—AUC-ROC (score: 84.28%). The usage of the transfer-learning technique and traditional deep learning engineering techniques was shown to enable us to obtain such results on consumer-class GPUs (graphics processing units).

18 September 2023 in Polish Journal of Radiology
Verified
Journal Article
Open access
Jakub KufelJakub Kufel
Jakub Kufel
IPIga Paszkiewicz
Iga Paszkiewicz
MBMichał Bielówka
Michał Bielówka
WBWiktoria Bartnikowska
Wiktoria Bartnikowska
MJMichał Janik
Michał Janik
SMSylwia Mielcarska
Sylwia Mielcarska
Purpose: Rapid development of artificial intelligence has aroused curiosity regarding its potential applications in medical field. The purpose of this article was to present the performance of ChatGPT, a state-of-the-art language model in relation to pass rate of national specialty examination (PES) in radiology and imaging diagnostics within Polish education system. Additionally, the study aimed to identify the strengths and limitations of the model through a detailed analysis of issues raised by exam questions.Material and methods: he present study utilized a PES exam consisting of 120 questions, provided by Medical Exami­nations Center in Lodz. Questions were administered using openai.com platform that grants free access to GPT-3.5 model. All questions were categorized according to Bloom’s taxonomy to assess their complexity and difficulty. Following the answer to each exam question, ChatGPT was asked to rate its confidence on a scale of 1 to 5 to evaluate the accuracy of its response.Results: ChatGPT did not reach the pass rate threshold of PES exam (52%); however, it was close in certain question categories. No significant differences were observed in the percentage of correct answers across question types and sub-types.Conclusions: The performance of the ChatGPT model in the pass rate of PES exam in radiology and imaging diagnostics in Poland is yet to be determined, which requires further research on improved versions of ChatGPT.

8 September 2023 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
Jakub KufelJakub Kufel
Jakub Kufel
KBKatarzyna Bargieł-Łączek
Katarzyna Bargieł-Łączek
MKMaciej Koźlik
Maciej Koźlik
Łukasz CzogalikŁukasz Czogalik
Łukasz Czogalik
PDPiotr Dudek
Piotr Dudek
ZNZbigniew Nawrat
Zbigniew Nawrat
Diagnostic imaging has become an integral part of the healthcare system. In recent years, scientists around the world have been working on artificial intelligence-based tools that help in achieving better and faster diagnoses. Their accuracy is crucial for successful treatment, especially for imaging diagnostics. This study used a deep convolutional neural network to detect four categories of objects on digital chest X-ray images. The data were obtained from the publicly available National Institutes of Health (NIH) Chest X-ray (CXR) Dataset. In total, 112,120 CXRs from 30,805 patients were manually checked for foreign objects: vascular port, shoulder endoprosthesis, necklace, and implantable cardioverter-defibrillator (ICD). Then, they were annotated with the use of a computer program, and the necessary image preprocessing was performed, such as resizing, normalization, and cropping. The object detection model was trained using the You Only Look Once v8 architecture and the Ultralytics framework. The results showed not only that the obtained average precision of foreign object detection on the CXR was 0.815 but also that the model can be useful in detecting foreign objects on the CXR images. Models of this type may be used as a tool for specialists, in particular, with the growing popularity of radiology comes an increasing workload. We are optimistic that it could accelerate and facilitate the work to provide a faster diagnosis.

28 September 2022 in International Journal of Medical Sciences
Verified
Journal Article
Open access
PDF file
Jakub KufelJakub Kufel
Jakub Kufel
KBKatarzyna Bargieł
Katarzyna Bargieł
MKMaciej Koźlik
Maciej Koźlik
Łukasz CzogalikŁukasz Czogalik
Łukasz Czogalik
PDPiotr Dudek
Piotr Dudek
KGKatarzyna Gruszczyńska
Katarzyna Gruszczyńska
This systematic review focuses on using artificial intelligence (AI) to detect COVID-19 infection with the help of X-ray images.
1

Verified
Journal Article
Open access
PDF file
Jakub KufelJakub Kufel
Jakub Kufel
IPIga Paszkiewicz
Iga Paszkiewicz
SKSzymon Kocot
Szymon Kocot
ALAnna Lis
Anna Lis
PDPiotr Dudek
Piotr Dudek
ZNZbigniew Nawrat
Zbigniew Nawrat
Objectives: The purpose of this study is to evaluate the performance of our deep learning algorithm in calculating cardiothoracic ratio (CTR) and thus in the assessment of cardiomegaly or pericardial effusion occurrences on chest radiography (CXR). Methods: From a database of 8000 CXRs, 13 folders with a comparable number of images were created. Then, 1020 images were chosen randomly, in proportion to the number of images in each folder. Afterward, CTR was calculated using RadiAnt Digital Imaging and Communications in Medicine (DICOM) Viewer software (2023.1). Next, heart and lung anatomical areas were marked in 3D Slicer. From these data, we trained an AI model which segmented heart and lung anatomy and determined the CTR value. Results: Our model achieved an Intersection over Union metric of 88.28% for the augmented training subset and 83.06% for the validation subset. F1-score for subsets were accordingly 90.22% and 90.67%. In the comparative analysis of artificial intelligence (AI) vs. humans, significantly lower transverse thoracic diameter (TTD) (p < 0.001), transverse cardiac diameter (TCD) (p < 0.001), and CTR (p < 0.001) values obtained using the neural network were observed. Conclusions: Results confirm that there is a significant correlation between the measurements made by human observers and the neural network. After validation in clinical conditions, our method may be used as a screening test or advisory tool when a specialist is not available, especially on Intensive Care Units (ICUs) or Emergency Departments (ERs) where time plays a key role.