Список литературы

2518-1092

Научный результат. Информационные технологии

2518-1092

10.18413/2518-1092-2026-11-3-0-6

4359

ИСКУССТВЕННЫЙ ИНТЕЛЛЕКТ И ПРИНЯТИЕ РЕШЕНИЙ

<strong>ОЦЕНКА ВАЛИДНОСТИ ОБЩЕДОСТУПНЫХ НАБОРОВ ДАННЫХ МРТ ДЛЯ НЕЙРООНКОЛОГИЧЕСКОГО ОБНАРУЖЕНИЯ АНОМАЛИЙ: ВЫЯВЛЕНИЕ КРИТИЧЕСКИХ ПРОБЛЕМ КАЧЕСТВА ДАННЫХ</strong>

<strong>VALIDITY ASSESSMENT OF PUBLIC MRI DATASETS FOR NEURO-ONCOLOGICAL ABNORMALITY DETECTION: UNCOVERING CRITICAL DATA QUALITY ISSUES</strong>

Низамли

Яссер

Nizamli

Yasser

yanizamli@stud.etu.ru

Филатов

Антон Юрьевич

Filatov

Anton Yuryevich

aifilatov@etu.ru

Фадел

Веаам

Fadel

Weaam

vfadel@stud.etu.ru

Шичкина

Юлия Александровна

Shichkina

Yulia Alexandrovna

yashichkina@etu.ru

2026

11300

Надежность моделей машинного обучения в таких ответственных приложениях, как диагностика опухолей головного мозга, критически зависит от качества обучающих данных. В данной статье представлен анализ четырех широко используемых наборов данных МРТ головного мозга: Chakrabarty, Br35H, Bhuvaji и Figshare. Мы оцениваем эти наборы данных по ключевым критериям достоверности, включая происхождение данных, целостность меток, независимость выборки, репрезентативность и соблюдение этических норм. Наши результаты показывают, что только набор данных Figshare удовлетворяет большинству основных стандартов качества. Остальные три демонстрируют серьезные недостатки, такие как неправильная маркировка, скрытое дублирование и этические нарушения, которые ставят под угрозу их научную достоверность. Контролируемые эксперименты дополнительно демонстрируют, что эти проблемы качества данных могут искусственно завышать производительность модели на значительные величины, подрывая надежность эталонных данных. Мы приходим к выводу, что такие наборы данных, как Chakrabarty, Br35H и Bhuvaji, требуют существенной структурной доработки, прежде чем их можно будет ответственно использовать в клинической разработке ИИ.

The reliability of machine learning models in high-stakes applications like brain tumor diagnosis depends critically on the quality of training data. This paper presents a forensic audit of four widely used brain MRI datasets: Chakrabarty, Br35H, Bhuvaji, and Figshare. We evaluate these datasets against key validity criteria, including data provenance, label integrity, sample independence, representativeness, and ethical compliance. Our findings reveal that only the Figshare dataset satisfies most essential quality standards. The remaining three exhibit profound flaws – such as label inaccuracies, undisclosed duplication, and ethical lapses – that compromise their scientific validity. Controlled experiments further demonstrate that these data-quality issues can artificially inflate model performance by significant margins, undermining benchmark reliability. We conclude that datasets like Chakrabarty, Br35H, and Bhuvaji require substantial structural revision before they can be responsibly used in clinical AI development.

качество данныхограниченные данныеаномалии МРТнадежность моделиэтичный ИИ

data qualityconstrained dataMRI abnormalitiesmodel reliabilityethical AI

Список литературы

Manjunath R.V., Yashaswini Gowda N., Manu H.M., Bharath K.N., Manasa R. Brain tumor classification on MRI images using fine-tuned deep learning models. Machine Learning for Computational Science and Engineering. 1, 22. – 2025.

Md Jahid Hasan, Mahmudul Hasan, Sumya Akter, Abu Bakar Siddique Mahi, Md Palash Uddin. Enhancing brain tumor classification with a novel attention based explainable deep learning framework. Biomedical Signal Processing and Control. 112. – 2026.

Md Imran Nazir, Afsana Akter, Md Anwar Hussen Wadud, Md Ashraf Uddin. Utilizing customized CNN for brain tumor prediction with explainable AI. Heliyon. 10, 20 – 2024.

Tapan Kumar Das, Pradeep Kumar Roy, Mohy Uddin, Kathiravan Srinivasan, Chuan-Yu Chang, Shabbir Syed-Abdul. Early tumor diagnosis in brain MR images via deep convolutional neural network model. Computers, Materials & Continua. 68, 2. – 2021. – Р. 2413-2429.

Saran Raj Sowrirajan, Surendiran Balasubramanian. Brain tumor classification using machine learning and deep learning algorithms. International Journal of Electrical and Electronics Research. 10, 4. – 2022. – Р. 999-1004.

Nizamli Ya., Filatov A. MRI brain tumor classification using HOG features selected via impurity based importances measure. International Journal of Electrical and Electronics Research. 12, 4. – 2024. – Р. 1251-1257.

Yasin Kaya, Ezgisu Akat, Serdar Yıldırım. Fusion-Brain-Net: A novel deep fusion model for brain tumor classification. Brain and Behavior. 15, 5. – 2025.

S. Deepak P.M. Ameer. Brain tumor categorization from imbalanced MRI dataset using weighted loss and deep feature fusion. Neurocomputing. 520. – 2023. – Р. 94-102.

Sohaib Asif, Wenhui Yi, Qurrat Ul Ain, Jin Hou, Tao Yi, Jinhai Si. Improving effectiveness of different deep transfer learning-based models for detecting brain tumors from MR images. IEEE Access. 10. – 2022. – Р. 34716-34730.

Victor Oluwatosin Ajayi. A review on primary sources of data and secondary sources of Data. European Journal of Education and Pedagogy. 2, 3. – 2023.

Martin J. Willemink, Wojciech A. Koszek, Cailin Hardell, Jie Wu, Dominik Fleischmann, Hugh Harvey, Les R. Folio, Ronald M. Summers, Daniel L. Rubin, Matthew P. Lungren. Preparing medical imaging data for machine learning.  Radiology. 295, 1. – 2020. – Р. 4-15.

Robert Geirhos , Jörn-Henrik Jacobsen, Claudio Michaelis , Richard Zemel, Wieland Brendel, Matthias Bethge, Felix A. Wichmann. Shortcut learning in deep neural networks. Nature Machine Intelligence. 2. – 2020. – Р. 665–673.

Recht B., Roelofs R., Schmidt L., Shankar V. Do ImageNet classifiers generalize to ImageNet?. arXiv:1902.10811v2. - 2019.

Björn Barz, Joachim Denzler. Do we train on test data? purging CIFAR of near-duplicates. Journal of Imaging. 6, 6. – 2020.

Aghabagherloo, Alireza, Aydin Abadi, Sumanta Sarkar, Vishnu Asutosh Dasu, Bart Preneel. Impact of data duplication on deep neural network-based image classifiers: robust vs. standard models. arXiv:2504.00638v2. – 2025.

John R. Zech, Marcus A. Badgeley, Manway Liu, Anthony B. Costa, Joseph J. Titano, Eric Karl Oermann. Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study.  PLOS Medicine. 15, 11. – 2018.

W Nicholson 2nd Price, I Glenn Cohen. Privacy in the age of medical big data. Nature medicine. 25, 1. – 2019. – Р. 37-43.

Navoneel Chakrabarty. Brain MRI images for brain tumor detection. URL: https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection (accessed 13.08.2025).

Ahmed Hamada. Br35H: brain tumor detection 2020. URL: https://www.kaggle.com/datasets/ahmedhamada0/brain-tumor-detection (accessed 13.08.2025).

Sartaj Bhuvaji, Ankita Kadam, Prajakta Bhumkar, Sameer Dedge, Swati Kanchan. Brain tumor classification (MRI). URL: https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri (accessed 13.08.2025).

Jun Cheng. brain tumor dataset. URL: https://figshare.com/articles/dataset/brain_tumor_dataset/1512427/5 (accessed 10.07.2025).

Sajad Einy, Hasan Saygin, Hemrah Hivehch, Yahya Dorostkar Navaei. Local and deep features based convolutional neural network frameworks for brain MRI anomaly detection.  Complexity. – 2022. – Р. 11.

Isha Gupta, Swati Singh, Sheifali Gupta, Soumya Ranjan Nayak. Classification of brain tumours in MRI images using a convolutional neural network. Current Medical Imaging. 20. – 2023.

Zahid Ullah, Jihie Kim. Hierarchical deep feature fusion and ensemble learning for enhanced brain tumor MRI classification. Mathematics. 13, 17. – 2025.

Jaeyong Kang, Zahid Ullah, Jeonghwan Gwak. MRI-Based brain tumor classification using ensemble of deep features and machine learning classifiers. Sensors. 21, 6. – 2021.

Sonia Ben Brahim, Samia Dardouri, Ridha Bouallegue. Brain tumor detection using a deep CNN model. Applied Computational Intelligence and Soft Computing. – 2024. – Р. 15.

Neha Bhardwaj, Meenakshi Sood, Ss Gill. Design of transfer learning based deep CNN paradigm for brain tumor classification. WSEAS Transactions on Biology and Biomedicine. 21. – 2024. – Р. 162-169.

Oussama Bouguerra, Bilal Attallah, Youcef Brik. MRI-based brain tumor ensemble classification using two stage score level fusion and CNN models. Egyptian Informatics Journal. 28. – 2024.

Soheila Saeedi, Sorayya Rezayi, Hamidreza Keshavarz, Sharareh R. Niakan Kalhori. MRI-based brain tumor detection using convolutional deep learning methods and chosen machine learning techniques. BMC Medical Informatics and Decision Making. 23, 16. – 2023.

Parvathy Jyothi, Dhanasekaran Selvaraj. An improved xception based feature learning and honeybadger optimized long short-term memory for accurate brain tumor classification with dataset balancing approach. Journal of Theoretical and Applied Information Technology. 103, 9. – 2025.

Rajat Mehrotra, M. A. Ansari, Rajeev Agrawal, Md Belal Bin Heyat, Pragati Tripathi, Eram Sayeed, Saba Parveen, John Irish G. Lira, Hadaate Ullah. Deep convolutional network-based probabilistic selection approach for multiclassification of brain tumors using magnetic resonance imaging. International Journal of Intelligent Systems. 2025. – Р. 21.

Shakhnoza Muksimova, Sabina Umirzakova, Sevara Mardieva, Nargiza Iskhakova, Murodjon Sultanov, Young Im Cho. A lightweight attention-driven YOLOv5m model for improved brain tumor detection. Computers in Biology and Medicine. 188, – 2025.

Saif Ur Rehman Khan, Ming Zhao, Sohaib Asif, Xuehan Chen. Hybrid-NET: A fusion of DenseNet169 and advanced machine learning classifiers for enhanced brain tumor diagnosis. International Journal of Imaging Systems and Technology. 34, 1. – 2024.