A Case Study: Using Deep Learning for the Early Detection of Lung Cancer
Cancer, one of the most life-threatening disease kills around 9.6 million people every year. The cost of curing cancer is very high, according to statistics each chemotherapy session requires a whooping amount of 70,000 INR. In India, 70% of cancers are detected during its advanced stage. Around 63,000 people die to Lung Cancer every year and disturbingly this number is said to increase by five-fold by the end of 2025.
The cost of screening Lung Cancer is very high. According to a case study, A Male Patient 65 years of age, diagnosed with lung cancer with metastasis to the brain i.e. cancer had spread to the brain at the time of diagnosis. Just the cost of diagnostics i.e. CT Scans, PET Scans, MRIs for the brain, FNAC, biopsy, and other diagnostics came up to almost 1,00,000 INR.
More than 80% of lung cancer patients will survive for at least a year if diagnosed at the earliest stage compared to around 15% for people diagnosed with the most advanced stage of the disease. That being said, We must strive to detect Lung Cancer at its early stages before becomes fatal keeping reduced costs in mind. This is where Deep Learning comes into play, pre-trained models that help us categorize the severity and early detection of Lung Cancer by just taking the CT scan of the lungs with unmarked nodules.
Researchers at Stanford came up with a cheaper alternative of diagnosis called the CAD system, where the diagnosis follows the following pipeline: image preprocessing → detection of cancerous nodule candidates → nodule candidate false positive reduction → malignancy prediction for each nodule candidate → malignancy prediction for overall CT scan.
In conclusion, Machine Learning and Deep Learning have a great impact on society, wonders can be achieved with the same while making the world a better place by improving Health care, security, etc.
References:
