Diagnostic Performance of Generative Adversarial Network-Based Deep Learning Methods for Alzheimer’s Disease: A Systematic Review and Meta-Analysis

Alzheimer’s disease (AD) is the most common form of dementia. Currently, only symptomatic management is available, and early diagnosis and intervention are crucial for AD treatment. As a recent deep learning strategy, generative adversarial networks (GANs) are expected to benefit AD diagnosis, but their performance remains to be verified. This study provided a systematic review on the application of the GAN-based deep learning method in the diagnosis of AD and conducted a meta-analysis to evaluate its diagnostic performance. A search of the following electronic databases was performed by two researchers independently in August 2021: MEDLINE (PubMed), Cochrane Library, EMBASE, and Web of Science. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was applied to assess the quality of the included studies. The accuracy of the model applied in the diagnosis of AD was determined by calculating odds ratios (ORs) with 95% confidence intervals (CIs). A bivariate random-effects model was used to calculate the pooled sensitivity and specificity with their 95% CIs. Fourteen studies were included, 11 of which were included in the meta-analysis. The overall quality of the included studies was high according to the QUADAS-2 assessment. For the AD vs. cognitively normal (CN) classification, the GAN-based deep learning method exhibited better performance than the non-GAN method, with significantly higher accuracy (OR 1.425, 95% CI: 1.150–1.766, P = 0.001), pooled sensitivity (0.88 vs. 0.83), pooled specificity (0.93 vs. 0.89), and area under the curve (AUC) of the summary receiver operating characteristic curve (SROC) (0.96 vs. 0.93). For the progressing MCI (pMCI) vs. stable MCI (sMCI) classification, the GAN method exhibited no significant increase in the accuracy (OR 1.149, 95% CI: 0.878–1.505, P = 0.310) or the pooled sensitivity (0.66 vs. 0.66). The pooled specificity and AUC of the SROC in the GAN group were slightly higher than those in the non-GAN group (0.81 vs. 0.78 and 0.81 vs. 0.80, respectively). The present results suggested that the GAN-based deep learning method performed well in the task of AD vs. CN classification. However, the diagnostic performance of GAN in the task of pMCI vs. sMCI classification needs to be improved.

Systematic Review Registration: [PROSPERO], Identifier: [CRD42021275294].

Analysis of the screening and predicting characteristics of the house-tree-person drawing test for mental disorders: A systematic review and meta-analysis

BACKGROUND

The house-tree-person (HTP) drawing test has received growing attention from researchers as a common projective test. However, the methods used to select and interpret drawing indicators still lack uniformity.

OBJECTIVE

This study aims to integrate drawing indicators into the process of screening for or classifying mental disorders by conducting a systematic review and meta-analysis of the application of the HTP test.

METHODS

A search of the following electronic databases was performed in May 2022: PubMed, Web of Science, Embase, EBSCO, CNKI, VIP, and Wanfang. Screening and checking of the literature were performed independently by two researchers. The empirical studies published on the use of the HTP test in mental disorders and studies providing specific data on the occurrence frequency of drawing characteristics were analyzed. A total of 30 studies were included in the meta-analysis, including 665 independent effect sizes and 6,295 participants. The strength of the association between drawing characteristics of the HTP test and the prevalence of mental disorders was measured by the ratio (OR) with a 95% CI. Publication bias was assessed using a funnel plot, Rosenthal's fail-safe number (N _fs), and the trim and fill method.

RESULTS

The results revealed 50 drawing characteristics that appeared at least three times in previous studies, of which 39 were able to significantly predict mental disorders. The HTP test can be divided into the following four dimensions: house, tree, person, and the whole. These dimensions reflect the structure, size, and other characteristics of the picture. The results showed that the greatest predictor of mental disorders was the whole (OR = 4.20, p < 0.001), followed by the house (OR = 3.95, p < 0.001), the tree (OR = 2.70, p < 0.001), and the person (OR = 2.16, p < 0.001). The valid predictors can be categorized into the following four types: item absence, bizarre or twisted, excessive details, and small or simplified. The subgroup analysis showed that the affective-specific indicators included no motion, leaning house, and decorated roof; thought-specific indicators included excessive separation among items, no window, loss of facial features, and inappropriate body proportions; and common indicators of mental disorders included no additional decoration, simplified drawing, very small house, two-dimensional house, and very small tree.

CONCLUSION

These findings can promote the standardization of the HTP test and provide a theoretical reference for the screening and clinical diagnosis of mental disorders.