Search

Abstract 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.

Abstract Machine Learning (ML) is nowadays an essential tool in the analysis of Magnetic Resonance Imaging (MRI) data, in particular in the identification of brain correlates in neurological and neurodevelopmental disorders. ML requires datasets of appropriate size for training, which in neuroimaging are typically obtained collecting data from multiple acquisition centers. However, analyzing large multicentric datasets can introduce bias due to differences between acquisition centers. ComBat harmonization is commonly used to address batch effects, but it can lead to data leakage when the entire dataset is used to estimate model parameters. In this study, structural and functional MRI data from the Autism Brain Imaging Data Exchange (ABIDE) collection were used to classify subjects with Autism Spectrum Disorders (ASD) compared to Typical Developing controls (TD). We compared the classical approach (external harmonization) in which harmonization is performed before train/test split, with an harmonization calculated only on the train set (internal harmonization), and with the dataset with no harmonization. The results showed that harmonization using the whole dataset achieved higher discrimination performance, while non-harmonized data and harmonization using only the train set showed similar results, for both structural and connectivity features. We also showed that the higher performances of the external harmonization are not due to larger size of the sample for the estimation of the model and hence these improved performance with the entire dataset may be ascribed to data leakage. In order to prevent this leakage, it is recommended to define the harmonization model solely using the train set.

The investigation of functional magnetic resonance imaging (fMRI) data with traditional machine learning (ML) and deep learning (DL) classifiers has been widely used to study autism spectrum disorders (ASDs). This condition is characterized by symptoms that affect the individual’s behavioral aspects and social relationships. Early diagnosis is crucial for intervention, but the complexity of ASD poses challenges for the development of effective treatments. This study compares traditional ML and DL classifiers in the analysis of tabular data, in particular, functional connectivity measures obtained from the time series of a public multicenter dataset, and evaluates whether the features that contribute most to the classification task vary depending on the classifier used. Specifically, Support Vector Machine (SVM) classifiers, with both linear and radial basis function (RBF) kernels, and Extreme Gradient Boosting (XGBoost) classifiers are compared against the TabNet classifier (a DL architecture customized for tabular data analysis) and a Multi Layer Perceptron (MLP). The findings suggest that DL classifiers may not be optimal for the type of data analyzed, as their performance trails behind that of standard classifiers. Among the latter, SVMs outperform the other classifiers with an AUC of around 75%, whereas the best performances of TabNet and MLP reach 65% and 71% at most, respectively. Furthermore, the analysis of the feature importance showed that the brain regions that contribute the most to the classification task are those primarily responsible for sensory and spatial perception, as well as attention modulation, which is known to be altered in ASDs.

Abstract Background The role of computed tomography (CT) in the diagnosis and characterization of coronavirus disease 2019 (COVID-19) pneumonia has been widely recognized. We evaluated the performance of a software for quantitative analysis of chest CT, the LungQuant system, by comparing its results with independent visual evaluations by a group of 14 clinical experts. The aim of this work is to evaluate the ability of the automated tool to extract quantitative information from lung CT, relevant for the design of a diagnosis support model. Methods LungQuant segments both the lungs and lesions associated with COVID-19 pneumonia (ground-glass opacities and consolidations) and computes derived quantities corresponding to qualitative characteristics used to clinically assess COVID-19 lesions. The comparison was carried out on 120 publicly available CT scans of patients affected by COVID-19 pneumonia. Scans were scored for four qualitative metrics: percentage of lung involvement, type of lesion, and two disease distribution scores. We evaluated the agreement between the LungQuant output and the visual assessments through receiver operating characteristics area under the curve (AUC) analysis and by fitting a nonlinear regression model. Results Despite the rather large heterogeneity in the qualitative labels assigned by the clinical experts for each metric, we found good agreement on the metrics compared to the LungQuant output. The AUC values obtained for the four qualitative metrics were 0.98, 0.85, 0.90, and 0.81. Conclusions Visual clinical evaluation could be complemented and supported by computer-aided quantification, whose values match the average evaluation of several independent clinical experts. Key points We conducted a multicenter evaluation of the deep learning-based LungQuant automated software. We translated qualitative assessments into quantifiable metrics to characterize coronavirus disease 2019 (COVID-19) pneumonia lesions. Comparing the software output to the clinical evaluations, results were satisfactory despite heterogeneity of the clinical evaluations. An automatic quantification tool may contribute to improve the clinical workflow of COVID-19 pneumonia.

Nowadays, MRI is focused on using ultra-high static magnetic fields (>7 T) to increase the signal-to-noise ratio. The use of high fields, on the other hand, requires novel technical solutions as well as more stringent design criteria for specific absorption rate levels, reducing radiative effect and coil resistance. In this paper, two flexible RF coils for 7 T human magnetic resonance, and 298 MHz ultra-high frequency operations were analyzed and characterized. Imaging of lower human limbs is regarded as a case study. The lumped element theory and subsequent numerical simulations were used to fine-tune the single-coil element and the dual-coil array design, respectively. Here, we demonstrate how the shape, size, configuration, and presence of the sample influence the coil performance. The penetration depth of the $B_{1}$ -field and the specific absorption rate values have been determined numerically using two numerical surface phantoms: saline and a multilayer human tissue. A preliminary study in the presence of a saline solution phantom has been carried out to develop and validate the dual-coil system. The frequency response of the dual-coil array was measured to assess its robustness when coupled to twelve human volunteers. We found that our design is robust to variations in the anatomical properties of the human thighs, and hence to coil bending. The presented approach can be useful for the implementation of flexible devices with high sensitivity levels and low specific absorption rate.

Morphological changes that may arise through a treatment course are probably one of the most significant sources of range uncertainty in proton therapy. Non-invasive in-vivo treatment monitoring is useful to increase treatment quality. The INSIDE in-beam Positron Emission Tomography (PET) scanner performs in-vivo range monitoring in proton and carbon therapy treatments at the National Center of Oncological Hadrontherapy (CNAO). It is currently in a clinical trial (ID: NCT03662373) and has acquired in-beam PET data during the treatment of various patients. In this work we analyze the in-beam PET (IB-PET) data of eight patients treated with proton therapy at CNAO. The goal of the analysis is twofold. First, we assess the level of experimental fluctuations in inter-fractional range differences (sensitivity) of the INSIDE PET system by studying patients without morphological changes. Second, we use the obtained results to see whether we can observe anomalously large range variations in patients where morphological changes have occurred. The sensitivity of the INSIDE IB-PET scanner was quantified as the standard deviation of the range difference distributions observed for six patients that did not show morphological changes. Inter-fractional range variations with respect to a reference distribution were estimated using the Most-Likely-Shift (MLS) method. To establish the efficacy of this method, we made a comparison with the Beam’s Eye View (BEV) method. For patients showing no morphological changes in the control CT the average range variation standard deviation was found to be 2.5 mm with the MLS method and 2.3 mm with the BEV method. On the other hand, for patients where some small anatomical changes occurred, we found larger standard deviation values. In these patients we evaluated where anomalous range differences were found and compared them with the CT. We found that the identified regions were mostly in agreement with the morphological changes seen in the CT scan.

Magnetic resonance imaging (MRI) nowadays plays an important role in the identification of brain underpinnings in a wide range of neuropsychiatric disorders, including Autism Spectrum Disorders (ASD). Characterizing the hallmarks in these pathologies is not a straightforward task and machine learning (ML) is certainly one of the most promising tools for addressing complex and non-linear problems. ML algorithms and, in particular, deep neural networks (DNNs), need large datasets in order to be properly trained and thus ensure generalization capabilities on new data. Large datasets can be obtained by collecting images from different centers, thus bringing unavoidable biases in the analysis due to differences in hardware and scanning protocols between different centers. In this work, we dealt with the issue of multicenter MRI data harmonization by comparing two different approaches: the analytical ComBat-GAM procedure, whose effectiveness is already documented in the literature, and an originally developed site-adversarial deep neural network (ad-DNN). The latter aims to perform a classification task while simultaneously searching for site-relevant patterns in order to make predictions free from site-related biases. As a case study, we implemented DNN and ad-DNN classifiers to distinguish subjects with ASD with respect to typical developing controls based on functional connectivity measures derived from data of the multicenter ABIDE collection. The classification performance of the proposed ad-DNN, measured in terms of the area under the ROC curve (AUC), achieved the value of AUC = 0.70±0.03, which is comparable to that obtained by a DNN on data harmonized according to the analytical procedure (AUC = 0.71±0.01). The relevant functional connectivity alterations identified by both procedures showed an agreement between each other and with the patterns of neuroanatomical alterations previously detected in the same cohort of subjects.

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.

Changes in brain structure asymmetry have been reported in autism spectrum disorder. Here the authors investigate this issue using a large-scale sample consisting of 54 data sets.

Magnetic resonance fingerprinting (MRF) is highly promising as a quantitative MRI technique due to its accuracy, robustness, and efficiency. Previous studies have found high repeatability and reproducibility of 2D MRF acquisitions in the brain. Here, we have extended our investigations to 3D MRF acquisitions covering the whole brain using spiral projection k-space trajectories.Our travelling head study acquired test/retest data from the brains of 12 healthy volunteers and 8 MRI systems (3 systems at 3 T and 5 at 1.5 T, all from a single vendor), using a study design not requiring all subjects to be scanned at all sites. The pulse sequence and reconstruction algorithm were the same for all acquisitions.After registration of the MRF-derived PD T1 and T2 maps to an anatomical atlas, coefficients of variation (CVs) were computed to assess test/retest repeatability and inter-site reproducibility in each voxel, while a General Linear Model (GLM) was used to determine the voxel-wise variability between all confounders, which included test/retest, subject, field strength and site.Our analysis demonstrated a high repeatability (CVs 0.7–1.3% for T1, 2.0–7.8% for T2, 1.4–2.5% for normalized PD) and reproducibility (CVs of 2.0–5.8% for T1, 7.4–10.2% for T2, 5.2–9.2% for normalized PD) in gray and white matter.Both repeatability and reproducibility improved when compared to similar experiments using 2D acquisitions. Three-dimensional MRF obtains highly repeatable and reproducible estimations of T1 and T2, supporting the translation of MRF-based fast quantitative imaging into clinical applications.

In particle therapy, the uncertainty of the delivered particle range during the patient irradiation limits the optimization of the treatment planning. Therefore, an in vivo treatment verification device is required, not only to improve the plan robustness, but also to detect significant interfractional morphological changes during the treatment itself. In this article, an effective and robust analysis to detect regions with a significant range discrepancy is proposed. This study relies on an in vivo treatment verification by means of in-beam Positron Emission Tomography (PET) and was carried out with the INSIDE system installed at the National Center of Oncological Hadrontherapy (CNAO) in Pavia, which is under clinical testing since July 2019. Patients affected by head-and-neck tumors treated with protons have been considered. First, in order to tune the analysis parameters, a Monte Carlo (MC) simulation was carried out to reproduce a patient who required a replanning because of significant morphological changes found during the treatment. Then, the developed approach was validated on the experimental measurements of three patients recruited for the INSIDE clinical trial (ClinicalTrials.gov ID: NCT03662373), showing the capability to estimate the treatment compliance with the prescription both when no morphological changes occurred and when a morphological change did occur, thus proving to be a promising tool for clinicians to detect variations in the patients treatments.

We propose and evaluate a procedure for the explainability of a breast density deep learning based classifier. A total of 1662 mammography exams labeled according to the BI-RADS categories of breast density was used. We built a residual Convolutional Neural Network, trained it and studied the responses of the model to input changes, such as different distributions of class labels in training and test sets and suitable image pre-processing. The aim was to identify the steps of the analysis with a relevant impact on the classifier performance and on the model explainability. We used the grad-CAM algorithm for CNN to produce saliency maps and computed the Spearman’s rank correlation between input images and saliency maps as a measure of explanation accuracy. We found that pre-processing is critical not only for accuracy, precision and recall of a model but also to have a reasonable explanation of the model itself. Our CNN reaches good performances compared to the state-of-art and it considers the dense pattern to make the classification. Saliency maps strongly correlate with the dense pattern. This work is a starting point towards the implementation of a standard framework to evaluate both CNN performances and the explainability of their predictions in medical image classification problems.

No univocal and reliable brain-based biomarkers have been detected to date in Autism Spectrum Disorders (ASD). Neuroimaging studies have consistently revealed alterations in brain structure and function of individuals with ASD; however, it remains difficult to ascertain the extent and localization of affected brain networks. In this context, the application of Machine Learning (ML) classification methods to neuroimaging data has the potential to contribute to a better distinction between subjects with ASD and typical development controls (TD). This study is focused on the analysis of resting-state fMRI data of individuals with ASD and matched TD, available within the ABIDE collection. To reduce the multiple sources of heterogeneity that impact on understanding the neural underpinnings of autistic condition, we selected a subgroup of 190 subjects (102 with ASD and 88 TD) according to the following criteria: male children (age range: 6.5–13 years); rs-fMRI data acquired with open eyes; data from the University sites that provided the largest number of scans (KKI, NYU, UCLA, UM). Connectivity values were evaluated as the linear correlation between pairs of time series of brain areas; then, a Linear kernel Support Vector Machine (L-SVM) classification, with an inter-site cross-validation scheme, was carried out. A permutation test was conducted to identify over-connectivity and under-connectivity alterations in the ASD group. The mean L-SVM classification performance, in terms of the area under the ROC curve (AUC), was 0.75 ± 0.05. The highest performance was obtained using data from KKI, NYU and UCLA sites in training and data from UM as testing set (AUC = 0.83). Specifically, stronger functional connectivity (FC) in ASD with respect to TD involve (p < 0.001) the angular gyrus with the precuneus in the right (R) hemisphere, and the R frontal operculum cortex with the pars opercularis of the left (L) inferior frontal gyrus. Weaker connections in ASD group with respect to TD are the intra-hemispheric R temporal fusiform cortex with the R hippocampus, and the L supramarginal gyrus with L planum polare. The results indicate that both under- and over-FC occurred in a selected cohort of ASD children relative to TD controls, and that these functional alterations are spread in different brain networks.

Autism spectrum disorders (ASDs) are a heterogeneous group of neurodevelopmental conditions characterized by impairments in social interaction and communication and restricted patterns of behavior, interests, and activities. Although the etiopathogenesis of idiopathic ASD has not been fully elucidated, compelling evidence suggests an interaction between genetic liability and environmental factors in producing early alterations of structural and functional brain development that are detectable by magnetic resonance imaging (MRI) at the group level. This work shows the results of a network-based approach to characterize not only variations in the values of the extracted features but also in their mutual relationships that might reflect underlying brain structural differences between autistic subjects and healthy controls. We applied a network-based analysis on sMRI data from the Autism Brain Imaging Data Exchange I (ABIDE-I) database, containing 419 features extracted with FreeSurfer software. Two networks were generated: one from subjects with autistic disorder (AUT) (DSM-IV-TR), and one from typically developing controls (TD), adopting a subsampling strategy to overcome class imbalance (235 AUT, 418 TD). We compared the distribution of several node centrality measures and observed significant inter-class differences in averaged centralities. Moreover, a single-node analysis allowed us to identify the most relevant features that distinguished the groups.