Problems of assessing the quality of computer analysis of medical images
REVIEWS
Abstract
Introduction. The analysis of medical images and the creation of systems to support medical decision — making are among the modern and currently developing areas in medicine. For practical healthcare, the task of automating diagnostic processes is one of the priorities in terms of increasing the availability and number of diagnostic procedures performed, which are currently not compensated by the increase in the number of qualified personnel needed to ensure proper interpretation of the results and, as a result, the provision of modern medical care. The review is devoted to the analysis of modern ideas about the stages of conducting and identifying problems in assessing the quality of computer recognition of medical images, determining the main directions of further research. Purpose — to determine the status of the issue and identify the main problems of quality assessment at the stages of computer recognition of medical images, to outline ways to solve them. Materials and methods. The analysis and systematization of scientific literature over the past 10 years has been carried out in the databases Pub Med, Scopus, Google Scholar, e LIBRARY. Results. Based on the review, it can be argued that at present a certain stage has been established when conducting computer analysis of medical images: preprocessing, segmentation and identification (classification). The analysis of scientific publications has shown a wide variety of methods used to classify objects in medical images, in the absence of a unified methodology for assessing their quality. Considering that the results of each sub-step affect the final classification results, it is necessary to identify objects with the desired characteristics in the image, determine stable relationships between the identified objects and, based on a system analysis, determine reference quality indicators for each stage of the analysis. Conclision. A certain sequence of stages has been established for carrying out computer analysis of medical images: preprocessing, segmentation and identification (classification). It has been determined that at each stage, both specific and non-specific methods are used in varying proportions. A wide variety of methods for classifying objects in medical images is demonstrated, with the lack of a unified methodology for assessing their quality. Considering that the results of each sub-stage influence the final classification results, it is necessary to identify objects with the desired characteristics in the image, determine stable relationships between the identified objects, and, based on a system analysis, determine reference quality indicators for each stage of the analysis.
References
1. Ansari Z., Mateenuddin Q., Abdullah A. Performance research on medical data classification using traditional and soft computing techniques. Int J Recent Technol Eng. 2019;8(2S3):990–995.
2. Rashed B.M., Popescu N. Critical analysis of the current medical image-based processing techniques for automatic disease evaluation: systematic literature review. Sensors. 2022;22(18):7065. https://doi.org/10.3390/s22187065.
3. Филимонова Е.С., Тарасенко С.Л., Дыхно Ю.А., Хлебникова Ф.Б. Оценка эффективности цитологической диагностики злокачественных новообразований легких. Сибирское медицинское обозрение. 2014;3(87):65–69.
4. Наркевич А.Н., Плотников Д.В., Виноградов К.А., Катаева А.В. Сравнение методов отбора признаков для идентификации объектов на цифровых изображениях микроскопических препаратов. Инженерный вестник Дона. 2018;(2):23.
5. Андронова Н.Е., Гребенюк П.Е., Чмутин А.М. Алгоритм и программная реализация управления оттеночным контрастом цифровых изображений. Инженерный вестник Дона. 2016;4(43):61.
6. Сорокин М.И. Методы распознавания образов на изображениях. Аллея науки. 2017;2(9):895–906.
7. Обухова Н.А., Мотыко А.А., Поздеев А.А. Цифровая обработка эндоскопических изображений для систем поддержки врачебных решений. Известия высших учебных заведений России. Радиоэлектроника. 2018;(6):54–65.
8. Gomariz M., Blaya S., Acebal P., Carretero L. Real-time UV-visible spectroscopy analysis of purple membrane-polyacrylamide film formation taking into account Fano line shapes and scattering. Plo S One. 2014;9(10):e110518. https://doi.org/10.1371/journal.pone.0110518.
9. De Puey E.G. Traditional gamma cameras are preferred. Journal of nuclear cardiology: official publication of the American Society of Nuclear Cardiology. 2016;23(4):795–802. https://doi.org/10.1007/s12350-016-0460-0.
10. Chetia R., Boruah S.M., Sahu P.P. Quantum image edge detection using improved Sobel mask based on NEQR. Quantum Information Processing. 2021;20:1–25.
11. Kong X., Chen Q., Wang J., Gu G., Wang P., Qian W., Ren K., Miao X. Inclinometer Assembly Error Calibration and Horizontal Image Correction in Photoelectric Measurement Systems. Sensors (Basel, Switzerland). 2018;18(1):248. https://doi.org/10.3390/s18010248.
12. Yang F., Han C., Bai B. X., Zhang C. Projected image correction technology research on autonomous-perception anisotropic surfaces. Applied Optics. 2018;57(2):283–294. https://doi.org/10.1364/AO.57.000283.
13. Balyan M.K. Object image correction using an X-ray dynamical diffraction Fraunhofer hologram. Journal of Synchrotron Radiation. 2014;21(Pt 2):449–451. https://doi.org/10.1107/S1600577513034802.
14. Hugonnet H., Lee M., Park Y. Optimizing illumination in three-dimensional deconvolution microscopy for accurate refractive index tomography. Optics Express. 2021;29(5):6293–6301. https://doi.org/10.1364/OE.412510.
15. Oran O.F., Klassen L.M., Serrai H., Menon R.S. Demonstration and suppression of respiration-related artifacts in Bloch-Siegert shift-based B1+ maps of the human brain. NMR in Biomedicine. 2020;33(7):e4299. https://doi.org/10.1002/nbm.4299.
16. Marchant T.E., Joshi K.D., Moore C.J. Accuracy of radiotherapy dose calculations based on cone-beam CT: comparison of deformable registration and image correction based methods. Physics in Medicine and Biology. 2018;63(6):065003. https://doi.org/10.1088/1361-6560/aab0f0.
17. Mozumder M., Hauptmann A., Nissila I., Arridge S.R., Tarvainen T.A Model-Based Iterative Learning Approach for Diffuse Optical Tomography. IEEE Transactions on Medical Imaging. 2022;41(5):1289–1299. https://doi.org/10.1109/TMI.2021.3136461.
18. Whitehead A.J., Kirkland N.J., Engler A.J. Atomic force microscopy for live-cell and hydrogel measurement. Methods in Molecular Biology (Clifton, N.J.). 2021;2299:217–226. https://doi.org/10.1007/978-1-0716-1382-5_16.
19. Coster A.D., Wichaidit C., Rajaram S., Altschuler S.J., Wu L.F. A simple image correction method for high-throughput microscopy. Nature Methods. 2014;11(6):602. https://doi.org/10.1038/nmeth.2971.
20. Yang W., Yuste R. Correction to: Holographic Imaging and Stimulation of Neural Circuits. Advances in Experimental Medicine and Biology. 2021;1293:C1–C2. https://doi.org/10.1007/978-981-15-8763-4_45.
21. Forsyth D., Rock J.J. Intrinsic image decomposition using paradigms. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2022;44(11):7624–7637. https://doi.org/10.1109/TPAMI.2021.3119551.
22. Ерусланов Р.В., Орехова М.Н., Дубровин В.Н. Сегментация изображений органов забрюшинного пространства по компьютерным томографическим изображениям на основе функции уровня. Компьютерная оптика. 2015;(4):592–599.
23. Perumal S., Thambusamy V. Preprocessing by contrast enhancement techniques for medical images. Int J Pure Appl Math. 2018;118:3681–3688.
24. Onim A., Musyoki S., Kihato P. Selection of optimal SINR threshold in fractional frequency reuse by comparing Otsu’s and entropy method. Heliyon. 2022;8(11):e11736. https://doi.org/10.1016/j.heliyon.2022.e11736.
25. Azeroual A., Afdel K. Fast image edge detection based on Faber Schauder Wavelet and Otsu Threshold. Heliyon. 2018;3(12):e00485. https://doi.org/10.1016/j.heliyon.2017.e00485.
26. Yang Y., Tian D., Wu B. A fast and reliable noise-resistant medical image segmentation and bias field correction model. Magnetic Resonance Imaging. 2018;54:15–31. https://doi.org/10.1016/j.mri.2018.06.015.
27. Khorram B., Yazdi M. A New optimized thresholding method using ant colony algorithm for MR brain image segmentation. Journal of Digital Imaging. 2019;32(1):162–174. https://doi.org/10.1007/s10278-018-0111-x.
28. Xiong F., Qi X., Nattel S., Comtois P. Wavelet analysis of cardiac optical mapping data. Computers in Biology and Medicine. 2015;65:243–255. https://doi.org/10.1016/j.compbiomed.2015.06.022.
29. Ou X., Pan W., Zhang X., Xiao P. Skin image retrieval using Gabor wavelet texture feature. International Journal of Cosmetic Science. 2016;38(6):607–614. https://doi.org/10.1111/ics.12332.
30. Liu Y., Saleh Z., Song Y., Chan M., Li X., Shi C., Qian X., Tang X. Novel Wavelet-Based Segmentation of Prostate CBCT Images with Implanted Calypso Transponders. International journal of Medical Physics, Clinical Engineering and Radiation Oncology. 2017;6(3):336–343. https://doi.org/10.4236/ijmpcero.2017.63030.
31. Maji P., Roy S. Rough-fuzzy clustering and unsupervised feature selection for wavelet based MR image segmentation. Plo S One. 2015;10(4):e0123677. https://doi.org/10.1371/journal.pone.0123677.
32. Li D., Zang P., Chai X., Cui Y., Li R., Xing L. Automatic multiorgan segmentation in CT images of the male pelvis using region-specific hierarchical appearance cluster models. Medical Physics. 2016;43(10):5426. https://doi.org/10.1118/1.4962468.
33. Uzümcü M., van der Geest R.J., Swingen C., Reiber J.H., Lelieveldt B.P. Time continuous tracking and segmentation of cardiovascular magnetic resonance images using multidimensional dynamic programming. Investigative Radiology. 2006;41(1):52–62. https://doi.org/10.1097/01.rli.0000194070.88432.24.
34. Birjandtalab J., Baran Pouyan M., Cogan D., Nourani M., Harvey J. Automated seizure detection using limited-channel EEG and non-linear dimension reduction. Computers in Biology and Medicine. 2017;82:49–58. https://doi.org/10.1016/j.compbiomed.2017.01.011.
35. Леонтьева Л.Н. Последовательный выбор признаков при восстановлении регрессии. Машинное обучение и анализ данных. 2012;(3):335–346.
36. Грицинская В.Л. Использование компьютерных технологий при проведении диспансеризации детского населения Республики Тыва. В мире научных открытий. 2017;(2):158–167.
37. Быкова В.В., Катаева А.В. Методы и средства анализа информативности признаков при обработке медицинских данных. Программные продукты и системы. 2016;(2):172–178.
38. Ваулин Г.Ф., Махматов О.В., Длужневская М.А. К вопросу отбора объектов на цифровых медицинских изображениях. Международный научно-исследовательский журнал. 2023;(4):130.
39. Lai Z., Deng H. Medical image classification based on deep features extracted by deep model and statistic feature fusion with multilayer perceptron. Computational Intelligence and Neuroscience. 2018;2018:2061516. https://doi.org/10.1155/2018/2061516.
40. Patankar A.M., Thorat S.S. Diagnosis of Ophthalmic Diseases in Fundus Image Using various Machine Learning Techniques. In Proceedings of the 2021 6th International Conference on Communication and Electronics Systems (ICCES), Coimbatre, India; 2021:1114–1118.
41. Гультяева Т.А., Попов А.А., Саутин А.С. Методы статистического обучения в задачах регрессии и классификации. Новосибирск: Новосибирский государственный технический университет; 2016.
42. Fuyong Xing, Yuanpu Xie, Hai Su, Fujun Liu, Lin Yang. Deep Learning in Microscopy Image Analysis: A Survey. IEEE Transactions on Neural Networks and Learning Systems. 2018;29(10):4550–4568. https://doi.org/10.1109/TNNLS.2017.2766168.
43. Gu Y., Shen M., Yang J., Yang G.Z. Reliable label-efficient learning for biomedical image recognition. IEEE Trans Biomed Eng. 2019;66(9):2423–2432. https://doi.org/10.1109/TBME.2018.2889915.
44. Wallner J., Schwaiger M., Hochegger K., Gsaxner C., Zemann W., Egger J. A review on multiplatform evaluations of semi-automatic open-source based image segmentation for cranio-maxillofacial surgery. Comput Methods Programs Biomed. 2019;182:105102. https://doi.org/10.1016/j.cmpb.2019.105102.
45. Rosen A.F.G., Roalf D.R., Ruparel K., Blake J., Seelaus K., Villa L.P., Ciric R., Cook P.A., Davatzikos C., Elliott M.A., Garcia de La Garza A., Gennatas E.D., Quarmley M., Schmitt J.E., Shinohara R.T., Tisdall M.D., Craddock R.C., Gur R.E., Gur R.C., Satterthwaite T.D. Quantitative assessment of structural image quality. Neuroimage. 2018;169:407–418. https://doi.org/10.1016/j.neuroimage.2017.12.059.
46. Schulze H.G., Rangan S., Piret J.M., Blades M.W., Turner R.F.B. Developing fully automated quality control methods for preprocessing raman spectra of biomedical and biological samples. Appl Spectrosc. 2018;72(9):1322–1340. https://doi.org/10.1177/0003702818778031.
47. Sagawa H., Itagaki K., Matsushita T., Miyati T. Evaluation of motion artifacts in brain magnetic resonance images using convolutional neural network-based prediction of full-reference image quality assessment metrics. J Med Imaging (Bellingham). 2022;9(1):015502. https://doi.org/10.1117/1.JMI.9.1.015502.
48. Jang J., Bang K., Jang H., Hwang D. Alzheimer's Disease Neuroimaging Initiative. Quality evaluation of no-reference MR images using multidirectional filters and image statistics. Magn Reson Med. 2018;80(3):914–924. https://doi.org/10.1002/mrm.27084.
49. Khan M.U., Luo M.R., Tian D. No-reference image quality metrics for color domain modified images. J Opt Soc Am A Opt Image Sci Vis. 2022;39(6):B65–B77. https://doi.org/10.1364/JOSAA.450595.
