Artificial intelligence (AI) has introduced a big chance, and also a challenge for physicians in their clinical practice. As a core technology, AI will play important roles in reshaping health care system in the near future. AI learns from good quality big data to develop new health care models. Understanding what big data research can do is helpful to predict the new world introduced by AI in the coming years.
In this talk, we will use our previous big data research experience based on Taiwan’s National Health Insurance Research Database (NHIRD) as examples to introduce the feasibility of big data approach in digestive cancer prevention and prediction. Based on NHIRD, we found early Helicobacter pylori eradication and regular use of non-steroidal anti-inflammatory drugs (NSAIDs) associated with reduced risk of gastric cancer (GCA). Gastroscopy within 5 years is helpful to reduce GCA patients’ mortality. We reported that antiviral therapy was effective as a 2nd chemoprevention agent to reduce hepatocellular carcinoma (HCC) risk in patients with hepatitis B. Antiviral therapy and NSAIDs were also an effective 3rd chemopreventive method to reduce HCC recurrence in patients with HBV and HCV-related HCC after liver resection or radiofrequency ablation (RFA). In addition, we found regular ultrasonography and higher RFA operator volume significantly associated with HCC mortality.
Big data approach can also be used for personalized risk prediction and finding the molecular signaling pathways to inhibit carcinogenesis. For patients with peptic ulcer diseases, we used NHIRD to construct a nomogram to predict each individual patient’s risk score and categorized patients’ risk in the following 2 years. We also used NHIRD to find the usefulness of metformin and statins to treat HCC and explored the related signaling pathways with in vitro and in vivo studies. Currently, we are conducting a nationwide multicenter randomized clinical trial to confirm the chemopreventive effect of statins in HCC.
In conclusion, big data study has extended our ability to predict and to prevent digestive cancers. It provides twilight for us to understand what AI can do in the new era. May we wish a disease-free century? |
