Machine Learning Applications to Diffuse Reflectance Spectroscopy in Optical Diagnosis; A Systematic Review

TL;DR

This systematic review analyzes the integration of machine learning with diffuse reflectance spectroscopy (DRS) for optical diagnosis across 77 studies. It highlights that ML can effectively classify tissue types and detect cancer using DRS signals, with ex-vivo data predominating and SVM/LDA being common choices, while extended wavelength ranges and transfer learning emerge as key trends. The authors identify major gaps, including insufficient patient-wise data stratification, limited in-vivo validation, varied preprocessing and reporting standards, and a need for explainable AI to ensure clinical safety. They argue for future work focused on rigorous in-vivo validation, standardized metrics and reproducibility, patient-stratified evaluation, and integration of explainable ML methods to translate DRS+ML into reliable clinical tools. Collectively, the findings suggest strong potential for ML-enabled DRS to support surgical guidance and tissue differentiation, provided methodological rigor and transparency improve.

Abstract

Diffuse Reflectance Spectroscopy has demonstrated a strong aptitude for identifying and differentiating biological tissues. However, the broadband and smooth nature of these signals require algorithmic processing, as they are often difficult for the human eye to distinguish. The implementation of machine learning models for this task has demonstrated high levels of diagnostic accuracies and led to a wide range of proposed methodologies for applications in various illnesses and conditions. In this systematic review, we summarise the state of the art of these applications, highlight current gaps in research and identify future directions. This review was conducted in accordance with the PRISMA guidelines. 77 studies were retrieved and in-depth analysis was conducted. It is concluded that diffuse reflectance spectroscopy and machine learning have strong potential for tissue differentiation in clinical applications, but more rigorous sample stratification in tandem with in-vivo validation and explainable algorithm development is required going forward.

Figures (7)

• Figure 1: PRISMA diagram of literature search.
• Figure 2: Geographical Representation of the country of origin of identified records.
• Figure 3: Frequency of the tissue and examination type.
• Figure 4: Frequency of the implemented machine learning models.
• Figure 5: Trends in the number of publications per year.