The invention provides a hospital real-time
navigation path planning method and
system based on
deep learning, and relates to the technical field of wisdom medices.The walking posture of a patient is analyzed in real time through a
deep learning network model, firstly, the patient with the normal walking posture and the patient with the abnormal walking posture are recognized and distinguished, and according to the actual environment condition of a hospital, the real-time
navigation path of the hospital is planned; according to the method, a walking
path line suitable for patient demands is planned, an optimal navigation walking path suitable for special
crowds such as elderly patients, pregnant women and handicapped people can be provided for the special
crowds such as the elderly patients, the pregnant women and the handicapped people, the problems of
walking difficulty and
cross infection of the special
crowds are solved, and the doctor-seeing experience satisfaction degree of the patients and the
hospital service efficiency are improved. Meanwhile, the walking posture of the patient is tracked, a secondary abnormity alarm mechanism is set, the path is re-planned in case of an
abnormal posture, and an alarm is given, so that the patients of special crowds are rescued at the first time, and the
service quality is further improved.