Leveraging Deep Learning for Designing Healthcare Analytics Heuristic for Diagnostics.

Area of Science:

• Healthcare Informatics
• Artificial Intelligence in Medicine
• Big Data Analytics

Background:

• The exponential growth of healthcare data necessitates advanced analytics for effective decision-making.
• Accurate and timely disease diagnosis, particularly for infectious diseases like Dengue Fever and Covid-19, is critical for reducing mortality.
• Challenges persist in the early diagnosis of conditions such as Diabetes and associated comorbidities.

Purpose of the Study:

• To propose and validate a novel hybrid deep learning technique for medical diagnostics.
• To leverage healthcare big data for improved disease prediction and patient management.
• To introduce the Louvain Mani-Hierarchical Fold Learning healthcare analytics approach.

Main Methods:

• Development of a hybrid deep learning technique: Louvain Mani-Hierarchical Fold Learning.
• Validation using real-time datasets including Dengue Fever, Covid-19, and other infectious diseases.
• Application on endocrine diagnostic data derived from a large cohort of patients.

Main Results:

• Demonstrated maximum 0.952 correlations between clusters using Spearman correlation on comorbidities data.
• Achieved predictive accuracies of 0.727, 0.701, and 0.203 for induced rules using Laplace evaluation.
• Successfully applied the algorithm to diverse datasets from multiple healthcare institutions.

Conclusions:

• The proposed Louvain Mani-Hierarchical Fold Learning algorithm shows significant potential for medical diagnostics in healthcare informatics.
• This hybrid deep learning approach can effectively analyze large-scale healthcare data for disease prediction.
• Further testing on extensive healthcare big data is recommended to explore its diagnostic capabilities.