Page 54 - AIH-2-3

Artificial Intelligence in Health AI in medical diagnostics: A multi-disease approach

3.6. Musculoskeletal medicine as small, biased datasets and ethical concerns related to

AI can uncover causes of knee pain often missed by doctors, corporate-driven AI initiatives highlight the need for
especially in underserved populations. By identifying further research and validation in this field.
46
pain contributors beyond visible radiographic findings, AI 3.13. Radiology
may help address disparities in diagnosis and treatment for
conditions like osteoarthritis. AI is being used to analyze medical images, such as
computed tomography (CT) and MRI scans, for detecting
3.7. Neurology diseases. It can provide benefits such as noise reduction,
AI is being explored for diagnosing and forecasting the enhanced image quality, and anatomical landmarking,
progression of Alzheimer’s disease through MRI data proving particularly useful in scenarios where human
52-68
17
and ML models, such as CNNs. Generative adversarial expertise is limited or the data are complex.
networks (GANs) have shown promise in improving early AI’s potential in these fields is vast, though challenges
diagnostic accuracy and prognosis predictions. 42,62 related to bias, validation, and integration into clinical
practice must still be addressed to realize its full impact.
3.8. Oncology The integration of AI into the healthcare industry is
AI has broad applications in cancer diagnosis, risk transforming how data are accessed, processed, and used to
stratification, and treatment personalization. Algorithms support clinical decision-making. Large health companies
have demonstrated superior accuracy to human merging allows for greater accessibility to vast amounts of
experts in breast cancer detection and prostate cancer health data, which serves as the foundation for AI-driven
identification. 50-56 AI is also being explored for grading the solutions. 11-66 As AI algorithms continue to evolve, they
aggressiveness of sarcomas and molecular characterizations enable more robust clinical decision support systems
of tumors. 58-62 However, challenges remain, such as a lack (CDSSs), adapting through the use of ML techniques.
of external validation and transparency in some AI studies, Companies in the healthcare sector are increasingly
raising concerns about scientific robustness. focused on big data, seeking opportunities in areas such as
data assessment, storage, management, and analysis. This
3.9. Ophthalmology industry focus is crucial for AI-powered innovations that

AI is aiding in the early detection of eye diseases, such enhance healthcare services and outcomes.
as diabetic retinopathy, and has received United States Several major companies are leading the development
(U.S.) Food and Drug Administration (FDA) approval of AI technologies in healthcare. IBM, through its Watson
for diagnosing specific eye conditions. 62-66 AI promises to Oncology platform, is working with institutions such as
improve diagnostic rates and efficiency in ophthalmic care. Memorial Sloan Kettering Cancer Center and Cleveland
3.10. Pathology Clinic to develop AI applications for cancer treatment.
Other collaborations include chronic disease treatment
AI is revolutionizing digital pathology by helping diagnose with CVS Health and drug development analysis with
cancers and predicting genetic mutations. AI tools can Johnson and Johnson. Microsoft’s Hanover project aims
analyze large-scale samples for diseases such as colorectal to predict the most effective cancer drug treatments, while
and breast cancer, improving efficiency and accuracy. 60-68 Google’s DeepMind works with the UK’s National Health
However, widespread implementation requires more Service to detect health risks and develop cancer detection
prospective studies to demonstrate AI’s clinical utility. algorithms. Tencent, Intel, and startups like Lumiata also
contribute to healthcare AI development, focusing on
3.11. Primary care
diagnostic services, medical imaging, and patient care
In primary care, AI is being used for decision-making, solutions. Neuralink, founded by Elon Musk, is pushing
predictive modeling, and analytics. 50-62 While examples the boundaries of neuroprosthetics with its brain chip
of AI’s clinical efficacy are limited, it has shown positive technology, which interfaces with neural pathways to treat
effects on treatment choices when integrated with physician conditions like paralysis. Digital consultant apps, such
decision-making processes. as Babylon Health, use AI to offer medical consultations
based on symptoms and medical history. These innovations
3.12. Psychiatry are supported by business models that target different
AI applications in psychiatry include predictive models user groups, including patients, healthcare providers, and
for diagnosis and treatment outcomes, as well as chatbots payers, offering solutions such as data connectivity and
for mental health support. 66-72 However, challenges such personalized treatment recommendations.

Volume 2 Issue 3 (2025) 48 doi: 10.36922/aih.5173