How Mesothelioma Patients Can Benefit From AI

Further, AI has significantly affected the medical field, proving its ability to analyze images for radiology and supporting doctors in detecting cancerous tumors. This is going to be especially beneficial for cancers that are historically harder to diagnose. In recent years, AI’s evolution has been directed to improving mesothelioma research and health care. Studies have confirmed AI’s role in preliminary diagnostic tests. For mesothelioma patients specifically, artificial intelligence may be a key component to an earlier diagnosis, leading to targeted and improved treatment, ultimately changing prognosis and life expectancy. 

On top of being a rare cancer, mesothelioma takes upwards of 10 to 20 years to manifest in the body. There are four stages of mesothelioma, and if diagnosed in the third stage, patients have an average life expectancy of 16 months. Mesothelioma is caused by exposure to asbestos, a fibrous and fragile mineral that can be inhaled. Oftentimes patients have symptoms that are delayed or misdiagnosed with lung cancer or other lung-related diseases. Mesothelioma continues to target a broad population through second and third-wave exposure to asbestos. Veterans, construction workers, engineers, and those in similar occupations have been the most vulnerable to asbestos exposure. 

AI is valuable to mesothelioma patients as it can provide a clear diagnosis and allow them to begin treatment immediately, whereas previously, attempts to treat patients were less effective because of its discovery in later cancer stages. AI offers hope for otherwise bleak life expectancies, and although AI is not the only solution, if it proves reliable, it can transform prognosis. Other cancers, such as lung and breast cancer, can also benefit from AI through early detection. Once detected, doctors can move forward with treatment before the tumor becomes aggressive or progresses more. 

In October of 2019, researchers at Owkin created a deep-learning program, MesoNet, to identify potential mesothelioma patients by scanning tissue samples and utilizing this technology to understand who would be most responsive to specific treatments. Immunotherapy, radiation therapy, and surgery are optimized when doctors can address cancer before it spreads to complicated and inoperable places. AI was similarly instrumental in pinpointing new biological tumor characteristics and connecting them to an effective prognosis. 

This model has also been verified by pathologists at the Centre Leon Berard cancer institute in Lyon, France. MesoNet is another development in AI for mesothelioma diagnosis, not only predicting patients but also “‘identify[ing] new biomarkers within the stromal regions of the tumor microenvironment that were predictive of survival.’” 

Prior to this, machine learning, a process that involves a “machine's ability to skim through data and find patterns, thus learning from the data and then applying it to problems to make informed decisions,” has been put into practice to advance AI. As AI has made an impact in radiology image analysis, this technology holds copious amounts of knowledge, being able to process images faster and with more accuracy than doctors. 

DeepMind and Google Health have collaborated to produce an AI algorithm that is more accurate at discerning early breast cancer than human radiologists. Another type of AI, an artificial immune system, can assist doctors in detecting malignant pleural mesothelioma. The study revealed a 97.74% accuracy performance and exceeded current algorithms for diagnosis. 

This research proves AI’s credibility; yet, it does not qualify AI as independent of human discretion and supervision. If AI can predict cancer, it will always be to augment oncology and medicine, never as a replacement. 

As with any disease, timing is significant. AI is unlikely to take over human judgment, however, it can help doctors in contributing to higher rates of cancer remission. If AI can be a tool in oncology, whether through supplementary image analysis scanning or by discovering prospective cancer patients in order to specialize treatment, it demonstrates a way for patients to receive optimal care.