Comparison of Machine Learning Algorithms in the Prediction of Hospitalized Patients with Schizophrenia

Alonso, Susel Góngora; Marques, Gonçalo; Agarwal, Deevyankar; Díez, Isabel de la Torre; Franco-Martín, Manuel

doi:10.3390/s22072517

Cited by 17 publications

(6 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alonso et al (2022) used a similar machine learning algorithmic approach to Schizophrenia diagnosis with Random Forest. While they achieved an accuracy an accuracy of 72.7%, an AUC of 0.796, a precision of 72.8%, an F1 score of 0.727, and a recall of 72.7%, we achieved an accuracy, AUC, precision, F1 score, and recall that were 19.7%, 13.6%, 22.7%, 21.2%, and 21.2% higher, respectively [30]. The results of these prior research studies show that our methods and proposed model presents greater and more promising results.…”

Section: Discussionmentioning

confidence: 70%

See 1 more Smart Citation

Using Electroencephalographic Signal Processing and Machine Learning Binary Classification to diagnose Schizophrenia

Desai

2023

Preprint

View full text Add to dashboard Cite

Electroencephalography (EEG) is an electrical activity measurement technique used to identify brain activity in Schizophrenic patients. Novel machine learning methods have emerged with useful applications for Schizophrenia classification. This research aims to compare the performance of several models post signal processing, such as Random Forest (RF), Support Vector Machine (SVM), Extra Trees (ET), and K-Nearest Neighbor (KNN), in the classification of healthy and Schizophrenic patients. The dataset used in this study contains 14 healthy and 14 Schizophrenic patients (n=28), with 17 channels and 2 reference electrodes designated to each patient, from the Nalecz Institute of Biocybernetics and Biomedical Engineering and the Institute of Psychiatry and Neurology in Warsaw, Poland. Signal processing feature extraction was performed using time-series or frequency-series Electroencephalographic data. The results suggest that Random Forest achieved the best performance metrics, achieving an accuracy, precision, recall, F1 score, and AUC of 92.4%, 95.5%, 95.5%, 0.939, and 0.932, respectively. These results propose that machine learning algorithms can be used to classify hospitalized patients who may have Schizophrenia, a useful supplement to additional clinical diagnosis performed by physicians.

show abstract

Section: Discussionmentioning

confidence: 70%

“…There is noise in the EEG signals to be considered. Noise in EEG signals stem from muscle, eye movement, and blinking [30]. Although our proposed model achieved successful metrics, no artifact removal technique was used to remove noise in each EEG signal.…”

Section: Discussionmentioning

confidence: 98%

Using Electroencephalographic Signal Processing and Machine Learning Binary Classification to diagnose Schizophrenia

Desai

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Besides, the recent work of Góngora‐Alonso et al. [24] includes a literature review of Artificial Intelligence based methods applied to schizophrenia detection, though not based of statistical significance analysis. In this context, it is worth mentioning the recent work by Sairamya et al.…”

Section: Introductionmentioning

confidence: 99%

“…Among its conclusions is the proved capability of these AI systems to identify subjects at high-risk of psychosis conversion and to differentiate schizophrenia from other disorders. Besides, the recent work of Góngora-Alonso et al [24] includes a literature review of Artificial Intelligence based methods applied to schizophrenia detection, though not based of statistical significance analysis. In this context, it is worth mentioning the recent work by Sairamya et al [25], where the relaxed local neighbour difference pattern (RLNDiP) technique is proposed and a combination of RLNDiP features from both time domain and time−frequency domain is used.…”

Section: Introductionmentioning

confidence: 99%

A discriminant analysis of the P3b wave with electroencephalogram by feature‐electrode pairs in schizophrenia diagnosis

Arribas

San‐José‐Revuelta

2023

IET Signal Processing

View full text Add to dashboard Cite

Schizophrenia is a disease that affects approximately 1% of the population. Its early accurate diagnosis is of vital importance to apply adequate therapy as soon as possible. We present a Statistical Discriminant Diagnosing (SDD) system that discriminates between healthy controls and subjects and that supports diagnosis by a medical professional. The system works with {feature, electrode} EEG pairs which are selected based on the statistical significance of the p‐values computed over the brain P3b wave. A bank of evoked potential pre‐processed and filtered EEG signals is recorded during an auditory odd‐ball (AOD) task and serves as input to the SDD system. These EEG signals comprise 20 features and 17 electrodes, both in time (t) and frequency (f) domain. The relevance of the Parieto‐Temporal region is shown, allowing us to identify highly discriminant {feature, electrode} pairs in the detection of schizophrenia, resulting lower p‐values in both Right and Left Hemispheres, as well as in Parieto‐Temporal EEG signals. See for instance, the {PSE, P4} pair, with p‐value = 0.00003 for (parametric) t Student and p‐value = 0.00019 for (nonparametric) U Mann‐Whitney tests, both under the 15 Hz cutoff frequency of a low pass EEG preprocessing filter. The relevance of this pair is in agreement with previously published related results. The proposed SDD system may provide the human expert (psychiatrist) with an objective complimentary information to help in the early diagnosis of schizophrenia.

show abstract

“…Additionally, research has shown that classification performance in t‐fMRI studies may be influenced by a range of demographic and clinical factors, including age, sex, illness stage, medication dosage, and the presence of psychotic symptoms 3 . Furthermore, classification outcomes can also be influenced by various ML‐related issues, such as the choice of classification algorithms, the selection of features, sample size, and the use of multiple sites 23–25 . Thus, it is crucial to take these factors into account to better understand their potential influence on t‐fMRI classification performance.…”

mentioning

confidence: 99%

Unveiling the potential of machine learning in schizophrenia diagnosis: A meta‐analytic study of task‐based neuroimaging data

Wang,

Yan,

Yang

et al. 2023

Psychiatry Clin Neurosci

View full text Add to dashboard Cite

The emergence of machine learning (ML) techniques has opened up new avenues for identifying biomarkers associated with schizophrenia (SCZ) using task‐related fMRI (t‐fMRI) designs. To evaluate the effectiveness of this approach, we conducted a comprehensive meta‐analysis of 31 t‐fMRI studies using a bivariate model. Our findings revealed a high overall sensitivity of 0.83 and specificity of 0.82 for t‐fMRI studies. Notably, neuropsychological domains modulated the classification performance, with selective attention demonstrating a significantly higher specificity than working memory (β = 0.98, z = 2.11, P = 0.04). Studies involving older, chronic patients with SCZ reported higher sensitivity (P <0.015) and specificity (P <0.001) than those involving younger, first‐episode patients or high‐risk individuals for psychosis. Additionally, we found that the severity of negative symptoms was positively associated with the specificity of the classification model (β = 7.19, z = 2.20, P = 0.03). Taken together, these results support the potential of using task‐based fMRI data in combination with machine learning techniques to identify biomarkers related to symptom outcomes in SCZ, providing a promising avenue for improving diagnostic accuracy and treatment efficacy. Future attempts to deploy ML classification should consider the factors of algorithm choice, data quality and quantity, as well as issues related to generalization.

show abstract

Comparison of Machine Learning Algorithms in the Prediction of Hospitalized Patients with Schizophrenia

Cited by 17 publications

References 60 publications

Using Electroencephalographic Signal Processing and Machine Learning Binary Classification to diagnose Schizophrenia

Using Electroencephalographic Signal Processing and Machine Learning Binary Classification to diagnose Schizophrenia

A discriminant analysis of the P3b wave with electroencephalogram by feature‐electrode pairs in schizophrenia diagnosis

Unveiling the potential of machine learning in schizophrenia diagnosis: A meta‐analytic study of task‐based neuroimaging data

Contact Info

Product

Resources

About