Научные статьиM.H.Sayidova, S.U.Pulatova2025
MODERN APPROACHES TO MEDICAL IMAGE PROCESSING USING COMPUTER VISION BASED ON ARTIFICIAL INTELLIGENCE
Allashkurov Dilshodbek Mansurovich · Abdullayev Abdulla Fayzulla ugli (Tashkent state university of economy. Uzbekistan) · Jumanazarov Javlon Shohnazar ugli (Tashkent state university of economy. Uzbekistan) · Muminbek Khayrullayev (Tashkent state university of economy. Uzbekistan)
Журнал инноваций нового века · 42 · 60– 88 · 2026
Аннотация
This article explores modern approaches to medical image processing based on artificial intelligence (AI) and computer vision technologies. It analyzes the importance of acquiring, analyzing, enhancing, segmenting, and using images from MRI, CT, X-ray, and ultrasound scans for real-time clinical decision- making. The study also examines the role and potential of deep learning, convolutional neural networks (CNNs), and transfer learning in medical diagnostics. Keywords: artificial intelligence, computer vision, medical image processing, deep learning, CNN, diagnostics, image segmentation, real-time analysis. Medical image processing, crucial for modern medicine, merges computational methods with clinical applications. It's about getting, refining, and studying pictures from things like MRIs, CT scans, and ultrasounds. The basics include getting the image, getting it ready, splitting it into parts, and showing it in a clear way. This makes sure the pictures are good enough to help doctors figure out what's wrong and plan treatments that work. Also, things like artificial intelligence (AI) and computer vision are getting better all the time, which means they're helping to make medical image analysis faster and more accurate, generally speaking. For example, AI can quickly spot problems and help with diagnoses, letting doctors make smart choices more quickly, in most cases. So, the main ideas behind medical image processing aren't just about making images look better; they also help doctors give the right care at the right time, empowering physicians in delivering precise and timely medical care[1]. Medical imaging, a field rapidly expanding, includes several different methods. Each has specific benefits that assist in both diagnosing and planning treatments. For example, X-rays offer key information about bone structure. Because of this, they are commonly the first step when assessing many conditions. Computed tomography, or https://scientific-jl.com/new 11 99 33 Volume–78_Issue-2_June-2025 JOURNAL OF NEW CENTURY INNOVATIONS CT, builds on this by producing more detailed cross-sectional images. These images provide richer assessments of complex anatomy. In contrast, magnetic resonance imaging (MRI) excels with its high contrast in soft tissues. This makes it extremely useful for evaluating neurological and musculoskeletal issues. Then there's ultrasound, a method that's both non-invasive and affordable. It uses sound waves to create images of internal organs. As such, it's a common choice in obstetrics and cardiology. Lastly, innovations in nuclear medicine—think positron emission tomography (PET) scans— allow visualization of metabolic activity. These developments really highlight how medical imaging improves both patient outcomes and research prospects. Image acquisition, the starting point in medical imaging, is really important because it makes sure any analysis, processing, and interpretation later on gives us correct clinical info. This is where we grab pictures of the body's structure and any signs of disease. We do this using different methods like X-rays, MRIs, CT scans, and ultrasounds. Each method uses different physical principles—like electromagnetic radiation, for example, or sound waves—and they're each designed for specific kinds of diagnoses. Now, good image acquisition means paying attention to things like resolution, contrast, and noise. These things affect the image quality and how useful it is for diagnosing problems. Plus, new technology lets us use AI to automatically tweak settings to make images look even better by reducing errors and making things clearer. As medicine uses more and more computer vision, knowing how image acquisition works becomes key. It helps doctors get useful info from images, which leads to better care through accurate diagnoses and treatment plans[2]. Medical image processing relies heavily on image enhancement to boost the clarity and usefulness of MRI, CT scans, and X-rays for diagnosis. These techniques include things like histogram equalization, contrast stretching, and noise reduction, all designed to make key parts of medical images more visible. Artificial intelligence helps drive these algorithms, which not only refine image quality but also help spot subtle issues that could otherwise be missed. For example, deep learning has become a useful method for improving image resolution and contrast, which leads to more accurate patient assessments. Moreover, real-time enhancement means that evaluations can happen right away during procedures, which significantly improves clinical outcomes. Generally speaking, combining image enhancement with AI-driven computer vision represents a major advance in creating accurate and reliable medical imaging solutions. Image segmentation serves a crucial purpose in boosting the utility of medical images. It essentially links the initial data to insights ready for clinical use. By classifying pixels or voxels into segments, thes
Доказательный фрагмент
«This article explores modern approaches to medical image processing based on artificial intelligence (AI) and computer vision technologies.»
Образец извлечения
Демо-превьюThis article explores modern approaches to medical image processing based on artificial intelligence (AI) and computer vision technologies. Полный текст публикации CiteX не хранит — доступ к первоисточнику открывается по DOI согласно лицензии издателя и выбранному плану.
- Цитирований
- 0
- Просмотров
- 0
- Тип
- Журнал
- Страна
- Узбекистан
- Ключевые слова
- artificial intelligence, computer vision, medical image processing, deep learning, Keywords: artificial intelligence, diagnostics, CNN, image segmentation, real-time analysis.
Смежные источники
Общество и инновацииKhosiyat MAMATKULOVA2024
Академические исследования в современной наукеKhudaiberdiyeva Arafat Israilovna, Ph.D Professor hudayberdiyevaarofat@gmail.com +12026
Научные статьиI.B.Yangiboyev, E.X.Berdanov +22025