An Integrative Artificial Intelligence Framework for the Diagnosis of Multiple Diseases in Clinical Settings

The rapid expansion of Electronic Health Records (EHRs), imaging modalities, and biomolecular data has created both opportunities and challenges for enhancing clinical diagnosis through Artificial Intelligence (AI). This study presents an integrative AI framework that utilizes multi-modal data fusion combining structured clinical data, medical imaging, and laboratory test results for the simultaneous diagnosis of multiple diseases within a real-world clinical environment. The proposed framework leverages deep learning models, including convolutional neural networks (CNNs) for image analysis and attention-based recurrent architectures for sequential clinical data, supported by feature-level fusion and decision-level ensemble strategies. Experiments were conducted on publicly available and institutional datasets, demonstrating superior performance in diagnostic accuracy, precision, and F1-score across conditions such as cardiovascular disease, diabetes, pneumonia, and lung cancer. The system is designed to support real-time clinical decision-making, while adhering to data privacy and fairness principles. This research highlights the potential of integrated AI solutions in streamlining multi-disease diagnosis and improving patient outcomes across diverse healthcare settings.