Machine Learning-Powered fNIRS Detection of Idiopathic Central Precocious Puberty via Prefrontal Cortex Activation
Zeying Li1, Lifang Jia2,3, Yingxue Zou2,4
1College of Precision Instruments and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China.
Insights
This study shows functional near-infrared spectroscopy (fNIRS) can noninvasively detect brain differences in children with central precocious puberty (CPP). Machine learning models accurately distinguish CPP patients from healthy children, paving the way for faster diagnostics.
Area of Science:
- Neuroscience
- Pediatric Endocrinology
- Biomedical Engineering
Background:
- Idiopathic central precocious puberty (ICPP) diagnosis currently relies on invasive and lengthy tests.
- Functional near-infrared spectroscopy (fNIRS) presents a noninvasive alternative, but underlying neural mechanisms in ICPP are not fully understood.
- Automated diagnostic tools are needed to reliably differentiate ICPP patients from healthy children.
Purpose of the Study:
- To investigate prefrontal cortex (PFC) hemodynamic responses in children with ICPP compared to controls.
- To develop a noninvasive diagnostic model for ICPP using fNIRS and machine learning.
- To elucidate the neural mechanisms of PFC activation during cognitive tasks in ICPP.
Main Methods:
- Acquired fNIRS data from 167 participants (82 ICPP, 85 normal) during a mental arithmetic task.
- Analyzed group and gender-specific PFC activation patterns using general linear models.
- Extracted hemodynamic signal features and employed a conditional denoising diffusion probabilistic model (C-DDPM) for data augmentation.
Main Results:
- Normal participants showed more extensive PFC activation than the ICPP group, with gender-specific differences observed.
- A decision tree classifier achieved 86.57% accuracy in distinguishing between groups using specific channel features.
- Incorporating C-DDPM-generated synthetic data enhanced classifier performance.
Conclusions:
- The study clarifies PFC activation mechanisms in ICPP and normative development during cognitive tasks.
- Machine learning effectively distinguishes children with ICPP from healthy controls using fNIRS data.
- This research supports the development of automated, noninvasive, and rapid diagnostic tools for ICPP.
Abstract:
Objective and Impact Statement: This study examines prefrontal cortex (PFC) hemodynamic responses in children with idiopathic central precocious puberty (ICPP) versus normals and constructs a noninvasive diagnostic model using functional near-infrared spectroscopy (fNIRS) augmented by machine learning. Introduction: Current ICPP diagnosis relies on invasive and time-consuming gonadotropin-releasing hormone stimulation tests. While fNIRS offers a noninvasive alternative, the neural mechanisms underlying ICPP remain unclear, and reliable automated diagnostic tools distinguishing patients from healthy peers are lacking. Methods: fNIRS data were acquired from 167 participants (82 ICPP and 85 normal) during a mental arithmetic (MA) task. General linear models and statistical tests were employed to analyze group and gender-specific activation patterns. Multidimensional features were extracted from hemodynamic signals, and a conditional denoising diffusion probabilistic model (C-DDPM) was introduced for data augmentation. Results: Analysis revealed gender-specific disparities, with the normal group exhibiting more extensive PFC activation than the ICPP group. In classification, a decision tree model using features from key negatively correlated channels achieved 86.57% accuracy. Notably, integrating C-DDPM-generated synthetic data further improved classifier performance metrics. Conclusion: The study elucidates the mechanisms of PFC activation in both normative and ICPP-affected cohorts during MA tasks and validates the effectiveness of machine learning in distinguishing between normal and ICPP children. This study provides a scientific basis for the development of automated, noninvasive rapid diagnostic tools for ICPP.
