
Loading, please wait...

Loading, please wait...

Schizophrenia remains a severe neuropsychiatric disorder that drastically impacts an individual's daily functioning. While clinical assessments are traditional, modern schizophrenia detection via EEG offers a more objective diagnostic pathway. Researchers have identified that abnormal asymmetry in neural activity reflects significant cognitive impairment. Consequently, this study explores how complexity measures in brain dynamics can automate the diagnostic process.
The research team utilized a hybrid approach combining EEG-based asymmetric entropy analysis with a CNN-LSTM classification model. Specifically, they extracted feature maps from frequency bands using approximate, sample, and spectral entropies. Furthermore, they processed these maps through a pre-trained Inception-V3 model. The results showed that approximate and sample entropies provided superior discrimination between patients and healthy subjects. Because of this high sensitivity, the model achieved a 94.11% accuracy in the delta band. In addition, the precision reached 100%.
Moreover, the classification model demonstrated high effectiveness by identifying functional alterations. Consequently, the integration of Long Short-Term Memory (LSTM) units allowed the system to account for temporal dependencies. Therefore, this technology facilitates more accurate detection of pathological conditions. Specifically, it targets the irregularities found in channel pairs. Thus, clinicians can rely on these automated measures for better diagnostic clarity. Nevertheless, further validation remains important for diverse populations.
In contrast to standard metrics, asymmetric entropy feature maps provide a detailed view of brain connectivity. For instance, these maps capture the specific functional alterations caused by schizophrenia. Because the model utilizes inter-channel asymmetries, it identifies subtle pathological signatures. Consequently, this AI-driven method reduces the reliance on subjective interpretation. Therefore, it paves the way for precision psychiatry in clinical practice.
EEG captures electrical activity and identifies abnormal neural asymmetries and complexity measures that are characteristic of schizophrenia.
The study reported a high classification accuracy of 94.11% and a precision of 100%, particularly within the delta frequency band.
Disclaimer: This content is for informational and educational purposes only. It is not intended as medical advice or a substitute for professional clinical judgment, diagnosis, or treatment. Always seek the advice of a qualified healthcare provider with any questions regarding a medical condition. Refer to the latest local and national guidelines for clinical practice.
References
1. Selvarasu L et al. EEG based detection of schizophrenia using asymmetry of entropy and CNN-LSTM model. Proc Inst Mech Eng H. 2026 Mar 10. doi: 10.1177/09544119261422821. PMID: 41807278.
2. Bao X et al. A systematic review of EEG based automated schizophrenia classification through machine learning and deep learning. PMC. 2023.

Read summarized clinical updates, watch expert medical content, and earn CME certifications right from your smartphone.


A new CNN-LSTM model using EEG entropy asymmetry detects schizophrenia with 94.11% accuracy, particularly in the delta frequency band....
7 months ago

A prospective study introduces a clinical and biomarker-based score incorporating faecal calprotectin to predict microscopic colitis with high accuracy (AUC 0.90), outperforming older risk scores and preventing missed diagnoses during colonoscopy in chronic watery diarrhoea.
Today

Delhi is undergoing a major healthcare overhaul with enhanced Ayushman Bharat insurance up to ₹10 lakh, increased MBBS and PG medical training seats, expanded primary network mandirs, new tertiary neuro-ICUs, and public-private reproductive technology initiatives to build a resilient, preventative healthcare model.
Today

Researchers have engineered an innovative CRISPR-Cas12a fluorescent diagnostic kit for clinical sST2 detection. By combining nanotechnology with enzymatic trans-cleavage, this platform enables rapid, ultrasensitive heart failure diagnosis and risk stratification, outperforming conventional laboratory ELISA methods.
Today

Non-union fractures occur in 2-5% of cases, causing significant disability and cost. Conservative modalities such as LIPUS, PEMF, and focused ESWT offer non-surgical healing options for defects up to 5-10 mm when adequate mechanical stability exists without infection.
Today

A Brazilian birth cohort reveals significant sex-specific associations between preterm birth and adolescent adiposity, underscoring the urgent need for targeted long-term cardiometabolic surveillance and tailored clinical monitoring.
Today