Background
According to a study conducted by
Features of the technology
This technology was developed based on NEC's state-of-the-art AI, including 2D/3D human pose estimation technology and abductive reasoning technology, as well as TMDU's medical knowledge. Based on images of a user taken with a smart device, the technology automatically estimates the human skeletal pose structure, evaluates the condition of each body part, estimates the cause of CLBP, and suggests an exercise program tailored to the symptoms.Features of the technology are as follows:
1. Highly accurate human pose estimation even for images taken from various angles
When an individual uses a smartphone or tablet device to take images of themself, the images are normally taken from various angles. As a result, conventional technologies using images may estimate distorted skeletal structure depending on the shooting angles, resulting in a loss of accuracy.To overcome this issue, NEC developed a 2D/3D human pose estimation technology that estimates skeletal structure while considering the shooting angles. This enables highly accurate human pose estimation, even for images taken from various angles.
2. Evaluate the condition of each body part with high precision from images
In order to find the cause of CLBP, physical therapists and other specialists observe each movement, such as forward bending, backward bending, and rotation, and evaluate the condition of each body part, such as insufficient, moderate, or excessive joint flexion, based on the relationship between body parts, such as the angle between the pelvis and thighs, and the relationship between body parts and back shape. Conventional technology evaluates the condition of each body part by calculating the relationship between body parts based on the human skeletal information estimated from images. However, the accuracy of the condition evaluation is limited because it does not take into account the relationship between the body parts and the shape of the back.NEC has developed a technology to precisely estimate the shape of the back from images, and by taking into account the relationship between body parts and back shape, it is now possible to evaluate the condition of each body part with the same high accuracy as an expert.
3. Rapidly estimate the cause of CLBP
In order to find the cause of CLBP, it is necessary to exhaustively search for physical issues from a kinematic perspective that could be the cause, such as "excessive lumbar flexion," based on information on each individual's attributes, such as age, gender and lifestyle, and observable information such as symptoms. However, this generates a huge number of combinations, and conventional inference techniques require several hours to search for the cause.NEC has developed a proprietary abductive reasoning technology using a Satisfiability Assessment Problem (SAT) solver. This enables rapid estimation of the major causes of chronic low back pain within an average of 10 seconds, based on image and interview data. In addition, based on TMDU's medical knowledge, a knowledge base has been created that includes the condition of each body part, interview data, and the causes of chronic back pain, enabling highly accurate and comprehensive estimation of the causes of an individual's chronic back pain.After the above 1-3, an exercise program suitable for improving CLBP is presented on a user's device according to the presumed cause of the pain. Since exercise programs are provided by video, individuals with CLBP can perform the exercise program at home by themselves.
Future developments
NEC and TMDU aim to verify the effectiveness of this technology through testing at an NEC care center in
About
About
Copyright 2024 JCN Newswire . All rights reserved.
© Japan Corporate News, source