НЕПАРАМЕТРИЧНИЙ ПIДХIД ДО ПОЯСНЮВАЛЬНОГО ШТУЧНОГО IНТЕЛЕКТУ ТА ЙОГО ЗАСТОСУВАННЯ В МЕДИЦИНI
DOI:
https://doi.org/10.17721/2706-9699.2023.2.02Ключові слова:
пояснюваний штучний iнтелект, непараметрична статистика, постулати машинного навчання, глибинне навчання, згорткова нейронна мережаАнотація
В роботi пропонується непараметричний пiдхiд до пояснюваного штучного iнтелекту на основi постулату компактностi, який стверджує, що об’єкти одного класу в просторi ознак, як правило, розташованi ближче один до одного, нiж до об’єктiв iнших класiв. Загальноприйнято вважати об’єкти подiбними, якщо вони розташованi близько в просторi ознак. Мiж тим, властивостi предметiв у реальному життi часто є випадковими значеннями. Такi об’єкти описуються не вектором ознак, а випадковою вибiркою або кiлькома вибiрками ознак, i постулат компактностi слiд замiнити постулатом статистичної однорiдностi. Об’єкти вважаються однорiдними, якщо їхнi ознаки пiдкоряються однаковим розподiлам. У роздiлi описується непараметрична мiра однорiдностi та надається iлюстрацiя їх використання в медичних додатках, зокрема для дiагностики раку молочної залози в рамках пояснюваного штучного iнтелекту на основi подiбностi. Для порiвняння наводяться результати дiагностики того ж самого набору даних за допомогою глибинного навчання штучної нейронної мережi. Ми формулюємо новi статистичнi постулати машинного навчання та пропонуємо вважати алгоритм машинного навчання пояснювальним та iнтерпретованим, якщо вiн задовольняє цим постулатам.
Посилання
Adadi A., Berrada M. Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI). IEEE Access 2018. Vol. 6. P. 52138–52160.
Alabdulhadi M., Coolen-Maturi T., Coolen F. Nonparametric predictive inference for comparison of two diagnostic tests. Communications in Statistics – Theory and Methods. 2021. Vol. 50. P. 4470–4486.
Amann J. et al. To explain or not to explain? – Artificial intelligence explainability in clinical decision support systems. PLOS Digital Health. 2022. Vol. 1(2). P. e0000016.
Andreichuk A. V., Boroday N. V., Golubeva K. M., Klyushin D. A. Artificial Intelligence System for Breast Cancer Screening Based on Malignancy-Associated
Changes in Buccal Epithelium. In: Enabling AI Applications in Data Science. Part of the Studies in Computational Intelligence book series (SCI, 2022, volume 911) Springer, 2022, pp. 267–285.
Arrieta A. B. et al. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities, and challenges toward responsible A. Information Fusion. 2020. Vol. 58. P. 82–115.
Bakalis E. et al. (2022) Universal Markers Unveil Metastatic Cancerous Cross-Sections at Nanoscale. Cancers. 2022. Vol. 14. No. 15. P. 3728.
Barnett V. (1976) The ordering of multivariate data. Journal of the Royal Statistical Society. Series A (General). Vol. 139. No. 3. P.318–355.
Bi W. et al. Artificial intelligence in cancer imaging: Clinical challenges and applications. CA Cancer Journal for Clinicians. 2019. Vol. 69. No. 2). P. 127–157.
Borys K. et al. Explainable AI in medical imaging: An overview for clinical practitioners – Saliency-based XAI approaches. European Journal of Radiology. 2023. Vol. 162. P. 110787.
Cascos I. (2007) Depth function as based of a number of observation of a random vector. Working Paper 07-29, Statistic and Econometric Series 07, 2:1–28.
Chaddad A, Peng J, Xu J, Bouridane A. Survey of Explainable AI Techniques in Healthcare. Sensors. 2023. Vol. 23. No. 2. P. 634.
Chen Z. et al. Artificial intelligence for assisting cancer diagnosis and treatment in the era of precision medicine. Cancer Communications 2021. Vol. 41. No. 11. P. 1100–1115.
Elkington L., Adhikari P., Pradhan P. Fractal Dimension Analysis to Detect the Progress of Cancer Using Transmission Optical Microscopy. Biophysica. 2022. Vol. 2. No. 1. P. 59–69.
Hacking S., Yakirevich E., Wang Y. From Immunohistochemistry to New Digital Ecosystems: A State-of-the-Art Biomarker Review for Precision Breast Cancer Medicine. Cancers. 2022. Vol. 14. No. 14. P. 3469.
Hill B. Posterior distribution of percentiles: Bayes’ theorem for sampling from a population. Journal of American Statistical Association. 1968. Vol. 63. P. 677–691.
Hill B. De Finetti’s theorem, induction, and A(n) or Bayesian nonparametric predictive inference (with discussion). In: D. V. Lindley, J. M. Bernardo, M. H. DeGroot, and A. F. M. Smith (Eds.), Bayesian statistics (1988, Vol. 3, pp. 211–241). Oxford: Oxford University Press.
Klyushin D. A., Petunin Yu. I. A Nonparametric Test for the Equivalence of Populations Based on a Measure of Proximity of Samples. Ukrainian Mathematical Journal. 2003. Vol. 55. No. 2. P. 181–198.
Klyushin D., Golubeva K.. Boroday N., Shervarly D. Breast cancer diagnosis using machine learning and fractal analysis of malignancy-associated changes in buccal epithelium. Chapter in: Artificial Intelligence, Machine Learning, and Data Science Technologies Future Impact and Well-Being for Society 5.0 N. Mohan, R. Singla, P. Kaushal, and S. Kadry (Eds.), CRC Press, 2021, P. 1–18.
Koopaie M., Kolahdooz S., Fatahzadeh M., Manifar S. Salivary biomarkers in breast cancer diagnosis: A systematic review and diagnostic meta-analysis. Cancer Medicine. 2022. Vol. 11. No. 13. P. 2644–2661.
Koshevoy G., Mosler K. Zonoid trimming for multivariate distributions. Annals of Statistics. 1997. Vol. 25. P. 1998–2017.
Li J., Du Q., Sun C. An improved box-counting method for image fractal dimension estimation. Pattern Recognition. Vol. 42. No. 11. P. 2460–2469.
Liang W. et al. Cancer cells corrupt normal epithelial cells through miR-let-7c-rich small extracellular vesicle-mediated downregulation of p53/PTEN. Intertational Journal of Oral Science. Vol. 14. No. 36.
Lipton Z. The mythos of model interpretability: In machine learning, the concept of interpretability is both important and slippery. ACM Queue. 2018. Vol. 16. No. 3. P. 31–57.
Liu R. J. On a notion of data depth based on random simplices. Annals of Statistics. 1990. Vol. 18. P. 405–414.
Love P. et al. Explainable Artificial Intelligence (XAI): Precepts, Methods, and Opportunities for Research in Construction. arXiv:2211.06579v2, 2022.
Lyashko S. Klyushin D., Alexeyenko V. Mulrivariate ranking using elliptical peeling. Cybernetic and Systems Analysis. 2013. Vol. 49. No. 4. P. 511–516.
Mosler K., Mozharovskyi P. Choosing among notions of multivariate depth statistics. Statistical Science. Vol. 37. No. 3. P. 348–368.
Mottl V., Seredin O., Krasotkina O. Compactness Hypothesis, Potential Functions, and Rectifying Linear Space in Machine Learning. In: International Conference Commemorating the 40th Anniversary of Emmanuil Braverman’s Decease, Boston, MA, USA, April 28-30, 2017, Invited Talks.
Nazir S., Dickson D., Akram M. Survey of explainable artificial intelligence techniques for biomedical imaging with deep neural networks. Computers in Biology and Medicine. 2023. Vol. 156. P. 106668.
Oja H. Descriptive statistics for multivariate distributions. Statistics and Probability Letters. 1983. Vol. 1. P. 327–332.
Petunin Yu., Rublev B. Pattern recognition using quadratic discriminant functions. Numerical and Applied Mathematics. 1996. Vol. 80. P. 89–104.
Polverini P., Nor F., Nor J. Crosstalk between cancer stem cells and the tumor microenvironment drives progression of premalignant oral epithelium. Frontiers in Oral Health. 2023. Vol. 3. No. 1095842.
Rudin C. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence. 2019. No. 1. P. 206–215.
Rudin C. et al. Interpretable machine learning: Fundamental principles and 10 grand challenges. Statistical Surveys. 2022. Vol. 16. P. 1-85.
Sanchez J., Martin-Landrove M. Morphological and Fractal Properties of Brain Tumors. Frontiers in Physiology. 2022. Vol. 13. No. 878391.
Sheu R.-K., Pardeshi M. A Survey on Medical Explainable AI (XAI): Recent Progress, Explainability Approach, Human Interaction and Scoring System. Sensors. 2022. Vol. 22. No. 8068.
Subramanian H. et al. Procedures for risk-stratification of lung cancer using buccal nanocytology. Biomedical Optics Express. (2016). Vol. 7. No. 9. P. 3795–3810.
Tukey J. Mathematics and the picturing of data. In: Proceedings of the International Congress of Mathematician (1975, pp. 523–531). Montreal, Canada.
Us-Krasovec M. et. al. Malignancy associated changes in epithelial cells of buccal mucosa: a potential cancer detection test. Analytical and Quantitative Cytology and Histology. 2005. Vol. 27. No. 5. P. 254–262.
Wu C. et al. Cancer-Associated Adipocytes and Breast Cancer: Intertwining in the Tumor Microenvironment and Challenges for Cancer Therapy. Cancers. 2023. Vol. 15. No. 3. P. 726.
Xu C. et al. Modeling and analysis fractal order cancer model with effects of chemotherapy. Chaos, Solitons and Fractals. 2022. Vol. 161. No. 112325.
Yang S., Folke T., Shafto P. A psychological theory of explainability. In: Proceedings of the 39th International Conference on Machine Learning, Baltimore, Maryland, USA, 2022, PMLR 162.
Zhang Y., Weng Y., Lund J. Applications of Explainable Artificial Intelligence in Diagnosis and Surgery. Diagnostics. 2022. Vol. 12. No. 2. P. 237.
Zuo Y., Serfling R. General notions of statistical depth function. Annals of Statistics. Vol. 28. P. 461–482.
