Artificial Intelligence in Technical Communication for Secure Healthcare: An Adaptive Framework for Android Malware Detection

Abstract

Objective:The broad uptake of Android based mobile health (mHealth) applications has, in a way, increased the risk of malware attacks, that can endanger private patient data and also disrupt secure technical messaging. This study now puts forward an Artificial Intelligence aided adaptive scaffold, meant for accurate detection that is scalable, plus more secure Android malware defense, specifically inside healthcare settings.
Material and Methods: So there was this Android malware dataset, with both benign and harmful applications, it was collected and then preprocessed in some standard way. After that, a kind of new Dot Hash Swordfish Algorithm was used, mainly to strengthen data confidentiality, with fast enough encryption/decryption. Then the decrypted data set was fed into a Deep Ensure AttackNet approach, meant for spotting malware. At the same time, an Adaptive PropAdam optimization method was added, and it kind of blends Adam and RMSProp together, so the feature learning improves, and it classifies each app as benign or malicious. Finally the whole thing was evaluated via accuracy, precision, recall , F1-score, ROC, computational time, and also scalability, to see how well it behaves under pressure.
Results: The proposed framework showed, in a kind of notable way, better malware detection performance by reaching 96.86% accuracy and an ROC score of 0.99. Its use of encryption together with deep learning and adaptive optimization helped ramp up classification accuracy, speed up convergence a bit, cut down on computational complexity, and also make it more scalable than the usual deep learning–driven malware detection approaches.
Conclusion: The proposed AI assisted framework efficiently strengthens cyber security for Android based healthcare apps, while it also helps with secure technical communication among healthcare professionals, patients, and the digital health platforms in general. Put together, the encryption layer, the smart malware detection, and that adaptive optimization approach makes it a kind of dependable and scalable solution, to shield sensitive healthcare data and also to improve the resilience of these newer digital healthcare systems.