Page 61 - AIH-2-3

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

before it occurs, demonstrate AI’s life-saving potential
by enabling healthcare providers to take preventive
measures. 18-38 Despite AI’s numerous advantages, it remains
a complementary tool to healthcare professionals rather
than a replacement. The human element in healthcare,
characterized by empathy, ethical judgment, and
experience, is indispensable. AI’s role in healthcare is to
support professionals by providing diagnostic suggestions
and real-time insights. However, ethical concerns
surrounding data privacy, algorithmic biases, and patient
consent must be addressed to ensure AI’s responsible and
effective use in personalized medicine. AI is reshaping the
healthcare landscape by improving diagnosis, treatment,
and personalized medicine. It enables healthcare providers Figure 2. Benefits of AI in medical diagnostics. The chart shows that
to offer more precise and timely interventions while among the listed benefits of AI in medical diagnostics, enhanced accuracy
reducing costs and increasing efficiency. As AI continues is perceived as the most significant, followed by early detection and
to evolve, collaboration between healthcare professionals predictive analytics, with personalized medicine being considered the
and AI will be critical in ensuring its ethical and effective least important of the four.
integration into medical practice. Looking ahead, the Abbreviation: AI: Artificial intelligence.
potential of AI to further enhance patient care is promising,
with ongoing research and development helping to unlock
even greater possibilities for the future of medicine.

8. Results and findings
The investigation highlights transformative advancements
in AI, particularly in DL and ML applications in healthcare,
with a focus on improving diagnostic accuracy, early
disease detection, and personalized care.
8.1. Crucial AI contributions in medical diagnostics
DL, powered by ANNs, has shown the most substantial
impact in medical diagnostics. Enhanced computational
resources, the availability of large, labeled datasets, and
accessible frameworks have propelled the success of DL,
particularly in medical imaging. The turning point for DL Figure 3. Challenges of AI in medical diagnostics
was marked by the ImageNet Large-Scale Visual Recognition Abbreviation: AI: Artificial intelligence.
Challenge (ILSVRC), where CNNs significantly reduced
error rates in object detection and classification tasks, timely intervention improves patient outcomes. For
surpassing traditional methods and, in some cases, human example, DL has been employed in medical imaging
performance. 24-28 Figures 2-5 demonstrate the research to differentiate bacterial pneumonia in pediatric chest
findings, showcasing the advancements that DL and ML radiographs and identify unique characteristics in chest CT
techniques have contributed to healthcare diagnostics. images, outperforming traditional diagnostic techniques.
These visualizations highlight DL’s effectiveness in analyzing In addition, hybrid models, including case-based
large datasets, detecting complex disease patterns, and reasoning (CBR) systems, have been used to diagnose skin
achieving high accuracy in disease prediction. diseases while ANN-based real-time monitoring systems
help patients manage critical health metrics, enhancing
8.2. Disease diagnosis and prediction through DL care during emergencies.
and ML
ML algorithms such as random forest, SVM, and
DL and ML models have shown high accuracy in logistic regression have also proven effective in disease
diagnosing critical diseases such as liver disease, heart prediction. In predicting type 2 diabetes (T2D), random
disease, Alzheimer’s disease, and various cancers. Early forest classifiers achieved high accuracy based on lifestyle
diagnosis is especially crucial in these diseases, where and health data, while mobile platforms leveraging random

Volume 2 Issue 3 (2025) 55 doi: 10.36922/aih.5173