IOT-ENHANCED PRECISION TREATMENT SYSTEM FOR LUNG CANCER PATIENTS USING DNA ANALYSIS AND MACHINE LEARNING ALGORITHMS

Abstract

Lung cancer is one of the major causes of death due to cancer in the whole world, and its treatment requires effective and flexible approaches in the medical field. The given research introduces the idea of creating an IoT-based precision treatment system combining real-time monitoring of vitals, genomic analysis, and machine learning to provide personal recommendations on therapy to the lung cancer patients. The system uses wearable sensors (MAX30102, DS18B20) and combines them with ESP8266 microcontroller to gather such physiological data as heart rate, oxygen saturation, and body temperature. At the same time, genomic data in public databases (Kaggle, DepMap, ResearchGate) is examined to find out potentially fatal mutations such as EGFR, ALK, KRAS, and ROS1, which allows targeted therapy to be selected. Several machine learning models Random Forest (RF), Support Vector Machines (SVM), Gradient Boosting (GB) and K-Nearest Neighbours (KNN) were trained to prescribe the best medication regimen on the basis of integrated physiological and genomic data. The RF model outperformed the other two with the highest accuracy of 99%, which is highly predictive in case of personalization of treatments. The main user interface provided by a mobile application, built with Flutter, connected to Firebase and cloud APIs, allows patients and clinicians to monitor real-time vitals and be reminded about their medication as well as modify a treatment plan accordingly. The suggested framework implies overcoming the drawbacks of conventional lung cancer treatment through constant monitoring and timely action as well as tailoring specific treatment to individuals, which may be more effective in terms of patient outcomes and minimization of adverse medication effects.