Explainable predictive modeling for limited spectral data

Akulich, Frantishek; Anahideh, Hadis; Sheyyab, Manaf; Dhananjay, Ambre,

doi:10.1016/j.chemolab.2022.104572

Cited by 19 publications

(4 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shapley Additive Explanations (SHAP) is considered a popular state-of-the-art approach to explaining machine learning models. As a result, SHAP calculates the contribution of each feature to the target value and the SHAP method can be used to analyze the prediction for both classification and regression models [ 21 ]. The SHAP value of each feature within the SVM model helps to select which feature is important and which bands are responsible for discrimination by the best algorithm between non-diabetic individuals and uncontrolled type 2 diabetic patients.…”

Section: Resultsmentioning

confidence: 99%

Salivary ATR-FTIR Spectroscopy Coupled with Support Vector Machine Classification for Screening of Type 2 Diabetes Mellitus

et al. 2023

View full text Add to dashboard Cite

The blood diagnosis of diabetes mellitus (DM) is highly accurate; however, it is an invasive, high-cost, and painful procedure. In this context, the combination of ATR-FTIR spectroscopy and machine learning techniques in other biological samples has been used as an alternative tool to develop a non-invasive, fast, inexpensive, and label-free diagnostic or screening platform for several diseases, including DM. In this study, we used the ATR-FTIR tool associated with linear discriminant analysis (LDA) and a support vector machine (SVM) classifier in order to identify changes in salivary components to be used as alternative biomarkers for the diagnosis of type 2 DM. The band area values of 2962 cm−1, 1641 cm−1, and 1073 cm−1 were higher in type 2 diabetic patients than in non-diabetic subjects. The best classification of salivary infrared spectra was by SVM, showing a sensitivity of 93.3% (42/45), specificity of 74% (17/23), and accuracy of 87% between non-diabetic subjects and uncontrolled type 2 DM patients. The SHAP features of infrared spectra indicate the main salivary vibrational modes of lipids and proteins that are responsible for discriminating DM patients. In summary, these data highlight the potential of ATR-FTIR platforms coupled with machine learning as a reagent-free, non-invasive, and highly sensitive tool for screening and monitoring diabetic patients.

show abstract

Section: Resultsmentioning

confidence: 99%

Salivary ATR-FTIR Spectroscopy Coupled with Support Vector Machine Classification for Screening of Type 2 Diabetes Mellitus

et al. 2023

View full text Add to dashboard Cite

show abstract

“…To complement the performance of the first model the Shapley Additive Explanations (SHAP) was obtained. The SHAP is an approach that calculates the contribution of each feature to the target value [Akulich et al 2022]. This SHAP value obtained for each feature helps to understand which bands in ATR-FTIR are the most important to discriminate non-diabetic individuals and type 2 diabetic patients.…”

Section: Interpretability Analysismentioning

confidence: 99%

Evolutionary Neural Architecture Search for Type 2 Diabetes Mellitus Diagnosis from Salivary ATR-FTIR Spectroscopy

Andrade,

Sabino-Silva,

Carneiro

2024

Anais Do XXIV Simpósio Brasileiro De Computação Aplicada À Saúde (SBCAS 2024)

View full text Add to dashboard Cite

The blood diagnosis of diabetes mellitus (DM) is accurate, but invasive. Attenuated Total Reflectance by Fourier Transform Infrared Spectroscopy (ATR-FTIR) is a green technology adopted in the detection of several diseases resulting in a non-invasive and accurate diagnosis. The analysis of ATR-FTIR data using deep learning techniques like Convolutional Neural Network (CNN) is promising. However, the challenges to find optimized architectures are barely explored in the ATR-FTIR literature. In this paper, we propose an Evolutionary Neural Architecture Search technique able to find optimized CNN architectures for salivary ATR-FTIR spectra for type 2 DM diagnosis using Genetic Algorithm as optimization approach.

show abstract

“…This study distinguishes between global and local as well as model-based and model-agnostic explanation methods. 22 This is attributed to the fact that features contributing to classification may not be those representative of a tissue from a biological point of view. As such, a feature may be considered important despite contributing to an incorrect classification.…”

Section: Explainabilitymentioning

confidence: 99%

Explainable machine learning for the multiclass classification of diffuse reflectance spectroscopy signals in orthopaedic applications

Rossberg,

Li,

Andersson-Engels

et al. 2024

Data Science for Photonics and Biophotonics

View full text Add to dashboard Cite

Revision total hip arthroplasty suffers from low visibility with intra-body navigation hinging primarily on auditory and tactile cues. Consequently, the risk of surgical injury increases. One proposition to increase surgical precision is integrating an algorithm which classifies encountered tissues based on their reflectance spectra into the surgical tools. Previous works have developed machine learning applications for the automatic, binary, classification of tissue based on diffuse reflectance spectroscopy (DRS) signals and exploratory investigations have successfully integrated DRS probes into surgical devices including surgical drills. However, one problem with these studies is a lack of transparency in the algorithms, which is important to increase practitioners' trust and prevent bias. This study developed four machine learning algorithms which simultaneously classified broadband DRS signals (355 -1850 nm) of six ovine tissue classes. The algorithms were Linear Discriminant Analysis (LDA), Random Forrest, Convolutional Neural Network (CNN), and a Transformer model. Class-wise wavelength importance was visualized using model-based methods to understand classification mechanisms and increase model-explainability. It is concluded that CNNs hold the potential for successful initial device design and medical integration.

show abstract

Explainable predictive modeling for limited spectral data

Cited by 19 publications

References 62 publications

Salivary ATR-FTIR Spectroscopy Coupled with Support Vector Machine Classification for Screening of Type 2 Diabetes Mellitus

Salivary ATR-FTIR Spectroscopy Coupled with Support Vector Machine Classification for Screening of Type 2 Diabetes Mellitus

Evolutionary Neural Architecture Search for Type 2 Diabetes Mellitus Diagnosis from Salivary ATR-FTIR Spectroscopy

Explainable machine learning for the multiclass classification of diffuse reflectance spectroscopy signals in orthopaedic applications

Contact Info

Product

Resources

About