# Accurate Estimation of Methemoglobin and Oxygen Saturation in Skin Tissue Using Diffuse Reflectance Spectroscopy and Artificial Intelligence

## TL;DR

This study introduces a noninvasive method using machine learning and diffuse reflectance spectroscopy to accurately estimate methemoglobin and oxygen saturation in skin tissue.

## Contribution

A novel machine learning model using artificial neural networks is developed to detect methemoglobin and oxygen saturation with high accuracy.

## Key findings

- The method achieved a mean absolute error of 0.0392% for methemoglobin concentration and 0.0273% for oxygen saturation.
- The technique was experimentally validated on human subjects, confirming its accuracy in detecting subtle changes in methemoglobin and hemoglobin levels.
- Broadband diffuse reflectance spectroscopy can noninvasively differentiate overlapping spectral features of methemoglobin and hemoglobin.

## Abstract

In this paper, we present a noninvasive method for the accurate estimation of methemoglobin concentration. The proposed technique incorporates a novel machine learning model using the artificial neural network to detect methemoglobin and oxygen saturation from the diffuse reflectance spectra of skin tissue. Sixty‐six spectra were simulated using a four‐layer tissue model with varying oxygen saturation and methemoglobin concentration. A multifiber probe‐based DRS setup in the visible and near‐infrared wavelength range was used. The best accuracy, with a mean absolute error (MAE) of 0.0392% for the concentration of methemoglobin and 0.0273% for the percentage of oxygen saturation on the created data set, was achieved. Our method was also experimentally verified using DRS spectra collected from human subjects. Consequently, the findings demonstrate the ability of broadband DRS to noninvasively differentiate subtle changes in methemoglobin and hemoglobin levels despite their overlapping spectral features.

Discover a noninvasive method for estimating methemoglobin levels in human tissue with remarkable precision. By leveraging a novel machine learning approach using artificial neural networks, this study showcases the potential of diffuse reflectance spectroscopy (DRS) to accurately detect methemoglobin and oxygen saturation. Experimentally validated on human subjects, the method achieves unprecedented accuracy, paving the way for advanced diagnostics and real‐time health monitoring without the need for invasive procedures.