Abstract: Optical arrays sensor has become the main tool for disease diagnosis due to the low cost, high detection speed and high accuracy. With the rapid development of artificial intelligence, the fusion application of machine learning and optical arrays sensor has brought new breakthroughs in the field of disease diagnosis. Machine learning can process the multi-dimensional data collected by the optical sensors and mine the relationship between the optical data and the disease types, which can improve the detection performance of sensors. The basic principle of machine learning algorithm and its advantages in optical arrays sensor are described, and the application status of machine learning in optical arrays sensor is analyzed. The detection method combining arrays sensor technology and machine learning will become the new trend for disease diagnosis in the future.