Dealing with confounders and outliers in classification medical studies: The Autism Spectrum Disorders case study

Automated diagnosis of autism with artificial intelligence: State of the art

Autism spectrum disorder (ASD) represents a panel of conditions that begin during the developmental period and result in impairments of personal, social, academic, or occupational functioning. Early diagnosis is directly related to a better prognosis. Unfortunately, the diagnosis of ASD requires a long and exhausting subjective process. We aimed to review the state of the art for automated autism diagnosis and recognition in this research. In February 2022, we searched multiple databases and sources of gray literature for eligible studies. We used an adapted version of the QUADAS-2 tool to assess the risk of bias in the studies. A brief report of the methods and results of each study is presented. Data were synthesized for each modality separately using the Split Component Synthesis (SCS) method. We assessed heterogeneity using the I 2 statistics and evaluated publication bias using trim and fill tests combined with ln DOR. Confidence in cumulative evidence was assessed using the GRADE approach for diagnostic studies. We included 344 studies from 186,020 participants (51,129 are estimated to be unique) for nine different modalities in this review, from which 232 reported sufficient data for meta-analysis. The area under the curve was in the range of 0.71-0.90 for all the modalities. The studies on EEG data provided the best accuracy, with the area under the curve ranging between 0.85 and 0.93. We found that the literature is rife with bias and methodological/reporting flaws. Recommendations are provided for future research to provide better studies and fill in the current knowledge gaps.

Deep learning based joint fusion approach to exploit anatomical and functional brain information in autism spectrum disorders

BACKGROUND

The integration of the information encoded in multiparametric MRI images can enhance the performance of machine-learning classifiers. In this study, we investigate whether the combination of structural and functional MRI might improve the performances of a deep learning (DL) model trained to discriminate subjects with Autism Spectrum Disorders (ASD) with respect to typically developing controls (TD).

MATERIAL AND METHODS

We analyzed both structural and functional MRI brain scans publicly available within the ABIDE I and II data collections. We considered 1383 male subjects with age between 5 and 40 years, including 680 subjects with ASD and 703 TD from 35 different acquisition sites. We extracted morphometric and functional brain features from MRI scans with the Freesurfer and the CPAC analysis packages, respectively. Then, due to the multisite nature of the dataset, we implemented a data harmonization protocol. The ASD vs. TD classification was carried out with a multiple-input DL model, consisting in a neural network which generates a fixed-length feature representation of the data of each modality (FR-NN), and a Dense Neural Network for classification (C-NN). Specifically, we implemented a joint fusion approach to multiple source data integration. The main advantage of the latter is that the loss is propagated back to the FR-NN during the training, thus creating informative feature representations for each data modality. Then, a C-NN, with a number of layers and neurons per layer to be optimized during the model training, performs the ASD-TD discrimination. The performance was evaluated by computing the Area under the Receiver Operating Characteristic curve within a nested 10-fold cross-validation. The brain features that drive the DL classification were identified by the SHAP explainability framework.

RESULTS

The AUC values of 0.66±0.05 and of 0.76±0.04 were obtained in the ASD vs. TD discrimination when only structural or functional features are considered, respectively. The joint fusion approach led to an AUC of 0.78±0.04. The set of structural and functional connectivity features identified as the most important for the two-class discrimination supports the idea that brain changes tend to occur in individuals with ASD in regions belonging to the Default Mode Network and to the Social Brain.

CONCLUSIONS

Our results demonstrate that the multimodal joint fusion approach outperforms the classification results obtained with data acquired by a single MRI modality as it efficiently exploits the complementarity of structural and functional brain information.

Quality assessment, variability and reproducibility of anatomical measurements derived from T1-weighted brain imaging: The RIN-Neuroimaging Network case study

Initiatives for the collection of harmonized MRI datasets are growing continuously, opening questions on the reliability of results obtained in multi-site contexts. Here we present the assessment of the brain anatomical variability of MRI-derived measurements obtained from T1-weighted images, acquired according to the Standard Operating Procedures, promoted by the RIN-Neuroimaging Network. A multicentric dataset composed of 77 brain T1w acquisitions of young healthy volunteers (mean age = 29.7 ± 5.0 years), collected in 15 sites with MRI scanners of three different vendors, was considered. Parallelly, a dataset of 7 "traveling" subjects, each undergoing three acquisitions with scanners from different vendors, was also used. Intra-site, intra-vendor, and inter-site variabilities were evaluated in terms of the percentage standard deviation of volumetric and cortical thickness measures. Image quality metrics such as contrast-to-noise and signal-to-noise ratio in gray and white matter were also assessed for all sites and vendors. The results showed a measured global variability that ranges from 11% to 19% for subcortical volumes and from 3% to 10% for cortical thicknesses. Univariate distributions of the normalized volumes of subcortical regions, as well as the distributions of the thickness of cortical parcels appeared to be significantly different among sites in 8 subcortical (out of 17) and 21 cortical (out of 68) regions of i nterest in the multicentric study. The Bland-Altman analysis on "traveling" brain measurements did not detect systematic scanner biases even though a multivariate classification approach was able to classify the scanner vendor from brain measures with an accuracy of 0.60 ± 0.14 (chance level 0.33).

Effect of Realistic Test Conditions on Perception of Speech, Music, and Binaural Cues in Normal-Hearing Listeners

PURPOSE

The purpose of this study was to determine the feasibility of online testing in a quiet room for three auditory perception experiments in normal-hearing listeners: speech, music, and binaural cue.

METHOD

Under Experiment 1, sentence perception was measured using fixed signal-to-noise ratios (SNRs: +10 dB, 0 dB, and -10 dB) and using adaptive speech reception threshold (SRT) procedures. The correct scores were compared between quiet room and soundproof booth listening environments. Experiment 2 was designed to compare melodic contour identification between the two listening environments. Melodic contour identification was assessed with 1, 2, and 4 semitone spacings. Under Experiment 3, interaural level difference (ILD) and interaural time differences (ITD) were measured as a function of carrier frequency. For both measures, two modulated tones (400-ms duration and 100-Hz modulation rate) were sequentially presented through headphones to both ears, and subjects were asked to indicate whether the sound moved to the left or right ear. The measured ITD and ILD were then compared between the two listening environments.

RESULTS

There were no significant differences in any outcome measures (SNR- and SRT-based speech perception, melodic contour identification, and ITD/ILD) between the two listening environments.

CONCLUSIONS

These results suggest that normal-hearing listeners may not require a controlled listening environment in any of the three auditory assessments. As comparable data can be obtained via the online testing tool, using the online auditory experiments is recommended.

AGGN: Attention-based glioma grading network with multi-scale feature extraction and multi-modal information fusion

In this paper, a magnetic resonance imaging (MRI) oriented novel attention-based glioma grading network (AGGN) is proposed. By applying the dual-domain attention mechanism, both channel and spatial information can be considered to assign weights, which benefits highlighting the key modalities and locations in the feature maps. Multi-branch convolution and pooling operations are applied in a multi-scale feature extraction module to separately obtain shallow and deep features on each modality, and a multi-modal information fusion module is adopted to sufficiently merge low-level detailed and high-level semantic features, which promotes the synergistic interaction among different modality information. The proposed AGGN is comprehensively evaluated through extensive experiments, and the results have demonstrated the effectiveness and superiority of the proposed AGGN in comparison to other advanced models, which also presents high generalization ability and strong robustness. In addition, even without the manually labeled tumor masks, AGGN can present considerable performance as other state-of-the-art algorithms, which alleviates the excessive reliance on supervised information in the end-to-end learning paradigm.

Improving the level of autism discrimination with augmented data by GraphRNN

Datasets are the key to deep learning in autism disease research. However, due to the small quantity and heterogeneity of samples in current public datasets, for example Autism Brain Imaging Data Exchange (ABIDE), the recognition research is not sufficiently effective. Previous studies primarily focused on optimizing feature selection methods and data augmentation to improve recognition accuracy. This research is based on the latter, which learns the edge distribution of a real brain network through the graph recurrent neural network (GraphRNN) and generates synthetic data that have an incentive effect on the discriminant model. Experimental results show that the synthetic data greatly improves the classification ability of the subsequent classifiers, for example, it can improve the classification accuracy of a 50-layer ResNet by up to 30% compared with the case without synthetic data.

Functional Connectome-Based Predictive Modeling in Autism

Autism is a heterogeneous neurodevelopmental condition, and functional magnetic resonance imaging-based studies have helped advance our understanding of its effects on brain network activity. We review how predictive modeling, using measures of functional connectivity and symptoms, has helped reveal key insights into this condition. We discuss how different prediction frameworks can further our understanding of the brain-based features that underlie complex autism symptomatology and consider how predictive models may be used in clinical settings. Throughout, we highlight aspects of study interpretation, such as data decay and sampling biases, that require consideration within the context of this condition. We close by suggesting exciting future directions for predictive modeling in autism.

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder that affects approximately 1% of the population and causes significant burdens. ASD's pathogenesis remains elusive; hence, diagnosis is based on a constellation of behaviors. Structural magnetic resonance imaging (sMRI) studies have shown several abnormalities in volumetric and geometric features of the autistic brain. However, inconsistent findings prevented most contributions from being translated into clinical practice. Establishing reliable biomarkers for ASD using sMRI is crucial for the correct diagnosis and treatment. In recent years, machine learning (ML) and specifically deep learning (DL) have quickly extended to almost every sector, notably in disease diagnosis. Thus, this has led to a shift and improvement in ASD diagnostic methods, fulfilling most clinical diagnostic requirements. However, ASD discovery remains difficult. This review examines the ML-based ASD diagnosis literature over the past 5 years. A literature-based taxonomy of the research landscape has been mapped, and the major aspects of this topic have been covered. First, we provide an overview of ML's general classification pipeline and the features of sMRI. Next, representative studies are highlighted and discussed in detail with respect to methods, and biomarkers. Finally, we highlight many common challenges and make recommendations for future directions. In short, the limited sample size was the main obstacle; Thus, comprehensive data sets and rigorous methods are necessary to check the generalizability of the results. ML technologies are expected to advance significantly in the coming years, contributing to the diagnosis of ASD and helping clinicians soon.

Multi-site harmonization of MRI data uncovers machine-learning discrimination capability in barely separable populations: An example from the ABIDE dataset

Machine Learning (ML) techniques have been widely used in Neuroimaging studies of Autism Spectrum Disorders (ASD) both to identify possible brain alterations related to this condition and to evaluate the predictive power of brain imaging modalities. The collection and public sharing of large imaging samples has favored an even greater diffusion of the use of ML-based analyses. However, multi-center data collections may suffer the batch effect, which, especially in case of Magnetic Resonance Imaging (MRI) studies, should be curated to avoid confounding effects for ML classifiers and masking biases. This is particularly important in the study of barely separable populations according to MRI data, such as subjects with ASD compared to controls with typical development (TD). Here, we show how the implementation of a harmo- nization protocol on brain structural features unlocks the case-control ML separation capability in the analysis of a multi-center MRI dataset. This effect is demonstrated on the ABIDE data collection, involving subjects encompassing a wide age range. After data harmonization, the overall ASD vs. TD discrimination capability by a Random Forest (RF) classifier improves from a very low performance (AUC = 0.58 ± 0.04) to a still low, but reasonably significant AUC = 0.67 ± 0.03. The performances of the RF classifier have been evaluated also in the age-specific subgroups of children, adolescents and adults, obtaining AUC = 0.62 ± 0.02, AUC = 0.65 ± 0.03 and AUC = 0.69 ± 0.06, respectively. Specific and consistent patterns of anatomical differences related to the ASD condition have been identified for the three different age subgroups.

Enhancing the impact of Artificial Intelligence in Medicine: A joint AIFM-INFN Italian initiative for a dedicated cloud-based computing infrastructure

Artificial Intelligence (AI) techniques have been implemented in the field of Medical Imaging for more than forty years. Medical Physicists, Clinicians and Computer Scientists have been collaborating since the beginning to realize software solutions to enhance the informative content of medical images, including AI-based support systems for image interpretation. Despite the recent massive progress in this field due to the current emphasis on Radiomics, Machine Learning and Deep Learning, there are still some barriers to overcome before these tools are fully integrated into the clinical workflows to finally enable a precision medicine approach to patients' care. Nowadays, as Medical Imaging has entered the Big Data era, innovative solutions to efficiently deal with huge amounts of data and to exploit large and distributed computing resources are urgently needed. In the framework of a collaboration agreement between the Italian Association of Medical Physicists (AIFM) and the National Institute for Nuclear Physics (INFN), we propose a model of an intensive computing infrastructure, especially suited for training AI models, equipped with secure storage systems, compliant with data protection regulation, which will accelerate the development and extensive validation of AI-based solutions in the Medical Imaging field of research. This solution can be developed and made operational by Physicists and Computer Scientists working on complementary fields of research in Physics, such as High Energy Physics and Medical Physics, who have all the necessary skills to tailor the AI-technology to the needs of the Medical Imaging community and to shorten the pathway towards the clinical applicability of AI-based decision support systems.