Fourier Transform Infrared Spectroscopy Based Complementary Diagnosis Tool for Autism Spectrum Disorder in Children...

Area of Science:

• Biochemistry
• Medical Diagnostics
• Spectroscopy

Background:

• Autism spectrum disorder (ASD) is a lifelong neurodevelopmental condition affecting 1-2% of children globally.
• Accurate early diagnosis is crucial, but auxiliary methods to complement clinical evaluation are lacking, increasing misidentification risks, especially in young children.

Purpose of the Study:

• To develop analytical models for the auxiliary diagnosis of ASD in children and adolescents.
• To utilize blood serum Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) spectra for ASD detection.
• To compare the efficacy of chemometric methods like Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA).

Main Methods:

• Development of analytical models using chemometrics (PCA and PLS-DA) on blood serum ATR-FTIR spectra.
• Infrared spectra served as predictor variables (X).
• Hierarchical clustering analysis was also performed for group discrimination.

Main Results:

• Both PCA and PLS-DA models demonstrated excellent classification performance between ASD individuals and healthy controls.
• The unsupervised PCA model achieved performance comparable to the supervised PLS-DA model.
• Hierarchical clustering analysis successfully discriminated between the two groups.

Conclusions:

• The developed PCA-based model is an effective and economical auxiliary tool for ASD diagnosis in clinical settings.
• ATR-FTIR spectroscopy combined with chemometrics offers a promising approach for ASD auxiliary diagnostics.
• The study highlights the potential of spectral analysis for improving ASD identification accuracy.