Development and validation of a new method by MIR-FTIR and chemometrics for the early diagnosis of leprosy and evaluation of the treatment effect

Abstract

Objective

Develop a new method for diagnosing leprosy and monitoring the pharmacological treatment effect of patients.

Material and methods

Plasma samples from patients diagnosed with leprosy (n = 211) who had not yet received any pharmacological treatment were collected at a basic health unit in Brazil. After treatment, samples were collected from the same patients (n = 125). Plasma samples from healthy volunteers were also collected (n = 179) and used as a control group. All samples were analyzed by Fourier transform mid-infrared spectrophotometry (MIR-FTIR). The spectral data of the samples were subjected to chemometric analysis. Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) were used to predict diagnosis and monitor pharmacological treatment.

Results

The PCA model successfully distinguished among three sample classes: healthy individuals, pre-treatment leprosy patients, and post-treatment leprosy patients. The PLS-DA algorithm accurately classified healthy, treated, and diseased samples, facilitating both reliable diagnosis and treatment monitoring for leprosy. The model achieved a sensitivity of 97 %–100 %, specificity of 100 %, and accuracy ranging from 99 % to 100 %. Furthermore, when tested on plasma samples from patients with other conditions—renal failure (n = 1032), hypertriglyceridemia (n = 100), hypercholesterolemia (n = 100), and mixed dyslipidemia (n = 100)—the model correctly classified these as negative for leprosy, with diagnostic specificity between 93 % and 96 %.

Conclusion

The MIR-FTIR technique combined with PLS-DA analysis proved to be a highly effective tool for screening leprosy patients and monitoring treatment outcomes. Given its low cost, this method could be easily implemented in laboratories across emerging and low-income countries.