SSPNet: An interpretable 3D-CNN for classification of schizophrenia using phase maps of resting-state complex-valued fMRI data

Explainability of three-dimensional convolutional neural networks for functional magnetic resonance imaging of Alzheimer's disease classification based on gradient-weighted class activation mapping

Currently, numerous studies focus on employing fMRI-based deep neural networks to diagnose neurological disorders such as Alzheimer's Disease (AD), yet only a handful have provided results regarding explainability. We address this gap by applying several prevalent explainability methods such as gradient-weighted class activation mapping (Grad-CAM) to an fMRI-based 3D-VGG16 network for AD diagnosis to improve the model's explainability. The aim is to explore the specific Region of Interest (ROI) of brain the model primarily focuses on when making predictions, as well as whether there are differences in these ROIs between AD and normal controls (NCs). First, we utilized multiple resting-state functional activity maps including ALFF, fALFF, ReHo, and VMHC to reduce the complexity of fMRI data, which differed from many studies that utilized raw fMRI data. Compared to methods utilizing raw fMRI data, this manual feature extraction approach may potentially alleviate the model's burden. Subsequently, 3D-VGG16 were employed for AD classification, where the final fully connected layers were replaced with a Global Average Pooling (GAP) layer, aimed at mitigating overfitting while preserving spatial information within the feature maps. The model achieved a maximum of 96.4% accuracy on the test set. Finally, several 3D CAM methods were employed to interpret the models. In the explainability results of the models with relatively high accuracy, the highlighted ROIs were primarily located in the precuneus and the hippocampus for AD subjects, while the models focused on the entire brain for NC. This supports current research on ROIs involved in AD. We believe that explaining deep learning models would not only provide support for existing research on brain disorders, but also offer important referential recommendations for the study of currently unknown etiologies.

EAG-RS: A Novel Explainability-Guided ROI-Selection Framework for ASD Diagnosis via Inter-Regional Relation Learning

Deep learning models based on resting-state functional magnetic resonance imaging (rs-fMRI) have been widely used to diagnose brain diseases, particularly autism spectrum disorder (ASD). Existing studies have leveraged the functional connectivity (FC) of rs-fMRI, achieving notable classification performance. However, they have significant limitations, including the lack of adequate information while using linear low-order FC as inputs to the model, not considering individual characteristics (i.e., different symptoms or varying stages of severity) among patients with ASD, and the non-explainability of the decision process. To cover these limitations, we propose a novel explainability-guided region of interest (ROI) selection (EAG-RS) framework that identifies non-linear high-order functional associations among brain regions by leveraging an explainable artificial intelligence technique and selects class-discriminative regions for brain disease identification. The proposed framework includes three steps: (i) inter-regional relation learning to estimate non-linear relations through random seed-based network masking, (ii) explainable connection-wise relevance score estimation to explore high-order relations between functional connections, and (iii) non-linear high-order FC-based diagnosis-informative ROI selection and classifier learning to identify ASD. We validated the effectiveness of our proposed method by conducting experiments using the Autism Brain Imaging Database Exchange (ABIDE) dataset, demonstrating that the proposed method outperforms other comparative methods in terms of various evaluation metrics. Furthermore, we qualitatively analyzed the selected ROIs and identified ASD subtypes linked to previous neuroscientific studies.

Dynamic functional network connectivity based on spatial source phase maps of complex-valued fMRI data: Application to schizophrenia

BACKGROUND

Dynamic spatial functional network connectivity (dsFNC) has shown advantages in detecting functional alterations impacted by mental disorders using magnitude-only fMRI data. However, complete fMRI data are complex-valued with unique and useful phase information.

METHODS

We propose dsFNC of spatial source phase (SSP) maps, derived from complex-valued fMRI data (named SSP-dsFNC), to capture the dynamics elicited by the phase. We compute mutual information for connectivity quantification, employ statistical analysis and Markov chains to assess dynamics, ultimately classifying schizophrenia patients (SZs) and healthy controls (HCs) based on connectivity variance and Markov chain state transitions across windows.

RESULTS

SSP-dsFNC yielded greater dynamics and more significant HC-SZ differences, due to the use of complete brain information from complex-valued fMRI data.

COMPARISON WITH EXISTING METHODS

Compared with magnitude-dsFNC, SSP-dsFNC detected additional and meaningful connections across windows (e.g., for right frontal parietal) and achieved 14.6% higher accuracy for classifying HCs and SZs.

CONCLUSIONS

This work provides new evidence about how SSP-dsFNC could be impacted by schizophrenia, and this information could be used to identify potential imaging biomarkers for psychotic diagnosis.

Dynamic functional network connectivity analysis in schizophrenia based on a spatiotemporal CPD framework

Objective.Dynamic functional network connectivity (dFNC), based on data-driven group independent component (IC) analysis, is an important avenue for investigating underlying patterns of certain brain diseases such as schizophrenia. Canonical polyadic decomposition (CPD) of a higher-way dynamic functional connectivity tensor, can offer an innovative spatiotemporal framework to accurately characterize potential dynamic spatial and temporal fluctuations. Since multi-subject dFNC data from sliding-window analysis are also naturally a higher-order tensor, we propose an innovative sparse and low-rank CPD (SLRCPD) for the three-way dFNC tensor to excavate significant dynamic spatiotemporal aberrant changes in schizophrenia.Approach.The proposed SLRCPD approach imposes two constraints. First, the L1regularization on spatial modules is applied to extract sparse but significant dynamic connectivity and avoid overfitting the model. Second, low-rank constraint is added on time-varying weights to enhance the temporal state clustering quality. Shared dynamic spatial modules, group-specific dynamic spatial modules and time-varying weights can be extracted by SLRCPD. The strength of connections within- and between-IC networks and connection contribution are proposed to inspect the spatial modules. K-means clustering and classification are further conducted to explore temporal group difference.Main results.82 subject resting-state functional magnetic resonance imaging (fMRI) dataset and opening Center for Biomedical Research Excellence (COBRE) schizophrenia dataset both containing schizophrenia patients (SZs) and healthy controls (HCs) were utilized in our work. Three typical dFNC patterns between different brain functional regions were obtained. Compared to the spatial modules of HCs, the aberrant connections among auditory network, somatomotor, visual, cognitive control and cerebellar networks in 82 subject dataset and COBRE dataset were detected. Four temporal states reveal significant differences between SZs and HCs for these two datasets. Additionally, the accuracy values for SZs and HCs classification based on time-varying weights are larger than 0.96.Significance.This study significantly excavates spatio-temporal patterns for schizophrenia disease.

Denoising brain networks using a fixed mathematical phase change in independent component analysis of magnitude-only fMRI data

Brain networks extracted by independent component analysis (ICA) from magnitude-only fMRI data are usually denoised using various amplitude-based thresholds. By contrast, spatial source phase (SSP) or the phase information of ICA brain networks extracted from complex-valued fMRI data, has provided a simple yet effective way to perform the denoising using a fixed phase change. In this work, we extend the approach to magnitude-only fMRI data to avoid testing various amplitude thresholds for denoising magnitude maps extracted by ICA, as most studies do not save the complex-valued data. The main idea is to generate a mathematical SSP map for a magnitude map using a mapping framework, and the mapping framework is built using complex-valued fMRI data with a known SSP map. Here we leverage the fact that the phase map derived from phase fMRI data has similar phase information to the SSP map. After verifying the use of the magnitude data of complex-valued fMRI, this framework is generalized to work with magnitude-only data, allowing use of our approach even without the availability of the corresponding phase fMRI datasets. We test the proposed method using both simulated and experimental fMRI data including complex-valued data from University of New Mexico and magnitude-only data from Human Connectome Project. The results provide evidence that the mathematical SSP denoising with a fixed phase change is effective for denoising spatial maps from magnitude-only fMRI data in terms of retaining more BOLD-related activity and fewer unwanted voxels, compared with amplitude-based thresholding. The proposed method provides a unified and efficient SSP approach to denoise ICA brain networks in fMRI data.

PKDN: Prior Knowledge Distillation Network for bronchoscopy diagnosis

Bronchoscopy plays a crucial role in diagnosing and treating lung diseases. The deep learning-based diagnostic system for bronchoscopic images can assist physicians in accurately and efficiently diagnosing lung diseases, enabling patients to undergo timely pathological examinations and receive appropriate treatment. However, the existing diagnostic methods overlook the utilization of prior knowledge of medical images, and the limited feature extraction capability hinders precise focus on lesion regions, consequently affecting the overall diagnostic effectiveness. To address these challenges, this paper proposes a prior knowledge distillation network (PKDN) for identifying lung diseases through bronchoscopic images. The proposed method extracts color and edge features from lesion images using the prior knowledge guidance module, and subsequently enhances spatial and channel features by employing the dynamic spatial attention module and gated channel attention module, respectively. Finally, the extracted features undergo refinement and self-regulation through feature distillation. Furthermore, decoupled distillation is implemented to balance the importance of target and non-target class distillation, thereby enhancing the diagnostic performance of the network. The effectiveness of the proposed method is validated on the bronchoscopic dataset provided by Harbin Medical University Cancer Hospital, which consists of 2,029 bronchoscopic images from 200 patients. Experimental results demonstrate that the proposed method achieves an accuracy of 94.78% and an AUC of 98.17%, outperforming other methods significantly in diagnostic performance. These results indicate that the computer-aided diagnostic system based on PKDN provides satisfactory accuracy in diagnosing lung diseases during bronchoscopy.

Explainable artificial intelligence (XAI) in radiology and nuclear medicine: a literature review

Rational

Deep learning (DL) has demonstrated a remarkable performance in diagnostic imaging for various diseases and modalities and therefore has a high potential to be used as a clinical tool. However, current practice shows low deployment of these algorithms in clinical practice, because DL algorithms lack transparency and trust due to their underlying black-box mechanism. For successful employment, explainable artificial intelligence (XAI) could be introduced to close the gap between the medical professionals and the DL algorithms. In this literature review, XAI methods available for magnetic resonance (MR), computed tomography (CT), and positron emission tomography (PET) imaging are discussed and future suggestions are made.

Methods

PubMed, Embase.com and Clarivate Analytics/Web of Science Core Collection were screened. Articles were considered eligible for inclusion if XAI was used (and well described) to describe the behavior of a DL model used in MR, CT and PET imaging.

Results

A total of 75 articles were included of which 54 and 17 articles described post and ad hoc XAI methods, respectively, and 4 articles described both XAI methods. Major variations in performance is seen between the methods. Overall, post hoc XAI lacks the ability to provide class-discriminative and target-specific explanation. Ad hoc XAI seems to tackle this because of its intrinsic ability to explain. However, quality control of the XAI methods is rarely applied and therefore systematic comparison between the methods is difficult.

Conclusion

There is currently no clear consensus on how XAI should be deployed in order to close the gap between medical professionals and DL algorithms for clinical implementation. We advocate for systematic technical and clinical quality assessment of XAI methods. Also, to ensure end-to-end unbiased and safe integration of XAI in clinical workflow, (anatomical) data minimization and quality control methods should be included.

Automated detection of schizophrenia using deep learning: a review for the last decade

Schizophrenia (SZ) is a devastating mental disorder that disrupts higher brain functions like thought, perception, etc., with a profound impact on the individual's life. Deep learning (DL) can detect SZ automatically by learning signal data characteristics hierarchically without the need for feature engineering associated with traditional machine learning. We performed a systematic review of DL models for SZ detection. Various deep models like long short-term memory, convolution neural networks, AlexNet, etc., and composite methods have been published based on electroencephalographic signals, and structural and/or functional magnetic resonance imaging acquired from SZ patients and healthy patients control subjects in diverse public and private datasets. The studies, the study datasets, and model methodologies are reported in detail. In addition, the challenges of DL models for SZ diagnosis and future works are discussed.

Towards greater neuroimaging classification transparency via the integration of explainability methods and confidence estimation approaches

The field of neuroimaging has increasingly sought to develop artificial intelligence-based models for neurological and neuropsychiatric disorder automated diagnosis and clinical decision support. However, if these models are to be implemented in a clinical setting, transparency will be vital. Two aspects of transparency are (1) confidence estimation and (2) explainability. Confidence estimation approaches indicate confidence in individual predictions. Explainability methods give insight into the importance of features to model predictions. In this study, we integrate confidence estimation and explainability approaches for the first time. We demonstrate their viability for schizophrenia diagnosis using resting state functional magnetic resonance imaging (rs-fMRI) dynamic functional network connectivity (dFNC) data. We compare two confidence estimation approaches: Monte Carlo dropout (MCD) and MC batch normalization (MCBN). We combine them with two gradient-based explainability approaches, saliency and layer-wise relevance propagation (LRP), and examine their effects upon explanations. We find that MCD often adversely affects model gradients, making it ill-suited for integration with gradient-based explainability methods. In contrast, MCBN does not affect model gradients. Additionally, we find many participant-level differences between regular explanations and the distributions of explanations for combined explainability and confidence estimation approaches. This suggests that a similar confidence estimation approach used in a clinical context with explanations only output for the regular model would likely not yield adequate explanations. We hope that our findings will provide a starting point for the integration of the two fields, provide useful guidance for future studies, and accelerate the development of transparent neuroimaging clinical decision support systems.

A Fair Performance Comparison between Complex-Valued and Real-Valued Neural Networks for Disease Detection

Our aim is to contribute to the classification of anomalous patterns in biosignals using this novel approach. We specifically focus on melanoma and heart murmurs. We use a comparative study of two convolution networks in the Complex and Real numerical domains. The idea is to obtain a powerful approach for building portable systems for early disease detection. Two similar algorithmic structures were chosen so that there is no bias determined by the number of parameters to train. Three clinical data sets, ISIC2017, PH2, and Pascal, were used to carry out the experiments. Mean comparison hypothesis tests were performed to ensure statistical objectivity in the conclusions. In all cases, complex-valued networks presented a superior performance for the Precision, Recall, F1 Score, Accuracy, and Specificity metrics in the detection of associated anomalies. The best complex number-based classifier obtained in the Receiving Operating Characteristic (ROC) space presents a Euclidean distance of 0.26127 with respect to the ideal classifier, as opposed to the best real number-based classifier, whose Euclidean distance to the ideal is 0.36022 for the same task of melanoma detection. The 27.46% superiority in this metric, as in the others reported in this work, suggests that complex-valued networks have a greater ability to extract features for more efficient discrimination in the dataset.