NCT04399811

Near-infrared Vision for Microcirculatory Status

Machine Learning-Based Near-infrared Vision to Evaluate the Microcirculatory of Critical Ill Patients: A Prospective Observational Study

• Status: Unknown
• Phase: —
• Study type: Observational
• Enrollment: 2,000 (estimated)
• Sponsor: Shanghai Zhongshan Hospital · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Not accepted

Summary

The investigators aimed to combine the image of near-infrared vision and machine learning method to evaluate the microcirculatory status of critical ill patients.

Detailed description

The heat distribution of body is determined by the circulatory status. The investigators plan to the near-infrared vision to collect heat distribution information of limbs. Then, the machine learning method will be performed to recognize the subtle differences between images. Due to lack of golden standard of microcirculatory status, indirect parameters (such as lactate clearance, capillary refill time) and clinical outcomes will be recorded to evaluate the performance of maching learning model.

Conditions

• Near-infrared Vision
• Machine Learning
• Microcirculatory Status

Interventions

Type	Name	Description
DIAGNOSTIC_TEST	Near-infrared Vision Photograph	The near-infrared image will be recorded (by a infrared camera connected to a laptop) for each enrolled patient

Timeline

• Start date: 2020-05-17
• Primary completion: 2023-12-31
• Completion: 2023-12-31
• First posted: 2020-05-22
• Last updated: 2020-05-22