Machine learning for classification and prediction of brain diseases: recent advances and upcoming challenges

Optimizing Machine Learning Models for Accessible Early Cognitive Impairment Prediction: A Novel Cost-effective Model Selection Algorithm

Cognitive impairment and dementia-related diseases develop several years before moderate or severe deterioration in cognitive function occurs. Nevertheless, most dementia cases, especially in low- and middle-income countries, remain undiagnosed because of limited access to affordable diagnostic tools. Additionally, the development of accessible tools for diagnosing and predicting cognitive impairment has not been extensively discussed in the literature. The objective of this study is to develop a cost-effective and highly accessible machine learning model to predict the risk of cognitive impairment for up to five years before clinical insight. We utilized easily accessible data from the National Alzheimer's Coordinating Center (NACC) Uniform Data Set (UDS) to train and evaluate various machine learning and deep learning models. A novel algorithm was developed to facilitate the selection of cost-effective models that offer high performance while minimizing development and operational costs. We conducted various assessments, including feature selection, time-series analyses, and external validation of the selected model. Our findings indicated that the Support Vector Machine (SVM) model was preferred over other high-performing neural network models because of its computational efficiency, achieving F2-scores of 0.828 in cross-validation and 0.750 in a generalizability test. Additionally, we found that demographic and historical health data are valuable for early prediction of cognitive impairment. This study demonstrates the potential of developing accessible solutions to predict cognitive impairment early using accurate and efficient machine learning models. Future interventions should consider creating cost-effective assessment tools to support global action plans and reduce the risk of cognitive impairment.

Explainable Machine Learning Models for Brain Diseases: Insights from a Systematic Review

In recent years, Artificial Intelligence (AI) methods, specifically Machine Learning (ML) models, have been providing outstanding results in different areas of knowledge, with the health area being one of its most impactful fields of application. However, to be applied reliably, these models must provide users with clear, simple, and transparent explanations about the medical decision-making process. This systematic review aims to investigate the use and application of explainability in ML models used in brain disease studies. A systematic search was conducted in three major bibliographic databases, Web of Science, Scopus, and PubMed, from January 2014 to December 2023. A total of 133 relevant studies were identified and analyzed out of a total of 682 found in the initial search, in which the explainability of ML models in the medical context was studied, identifying 11 ML models and 12 explainability techniques applied in the study of 20 brain diseases.

AI-based differential diagnosis of dementia etiologies on multimodal data

Differential diagnosis of dementia remains a challenge in neurology due to symptom overlap across etiologies, yet it is crucial for formulating early, personalized management strategies. Here, we present an artificial intelligence (AI) model that harnesses a broad array of data, including demographics, individual and family medical history, medication use, neuropsychological assessments, functional evaluations and multimodal neuroimaging, to identify the etiologies contributing to dementia in individuals. The study, drawing on 51,269 participants across 9 independent, geographically diverse datasets, facilitated the identification of 10 distinct dementia etiologies. It aligns diagnoses with similar management strategies, ensuring robust predictions even with incomplete data. Our model achieved a microaveraged area under the receiver operating characteristic curve (AUROC) of 0.94 in classifying individuals with normal cognition, mild cognitive impairment and dementia. Also, the microaveraged AUROC was 0.96 in differentiating the dementia etiologies. Our model demonstrated proficiency in addressing mixed dementia cases, with a mean AUROC of 0.78 for two co-occurring pathologies. In a randomly selected subset of 100 cases, the AUROC of neurologist assessments augmented by our AI model exceeded neurologist-only evaluations by 26.25%. Furthermore, our model predictions aligned with biomarker evidence and its associations with different proteinopathies were substantiated through postmortem findings. Our framework has the potential to be integrated as a screening tool for dementia in clinical settings and drug trials. Further prospective studies are needed to confirm its ability to improve patient care.

Deep Learning for Alzheimer's Disease Prediction: A Comprehensive Review

Alzheimer's disease (AD) is a neurological disorder that significantly impairs cognitive function, leading to memory loss and eventually death. AD progresses through three stages: early stage, mild cognitive impairment (MCI) (middle stage), and dementia. Early diagnosis of Alzheimer's disease is crucial and can improve survival rates among patients. Traditional methods for diagnosing AD through regular checkups and manual examinations are challenging. Advances in computer-aided diagnosis systems (CADs) have led to the development of various artificial intelligence and deep learning-based methods for rapid AD detection. This survey aims to explore the different modalities, feature extraction methods, datasets, machine learning techniques, and validation methods used in AD detection. We reviewed 116 relevant papers from repositories including Elsevier (45), IEEE (25), Springer (19), Wiley (6), PLOS One (5), MDPI (3), World Scientific (3), Frontiers (3), PeerJ (2), Hindawi (2), IO Press (1), and other multiple sources (2). The review is presented in tables for ease of reference, allowing readers to quickly grasp the key findings of each study. Additionally, this review addresses the challenges in the current literature and emphasizes the importance of interpretability and explainability in understanding deep learning model predictions. The primary goal is to assess existing techniques for AD identification and highlight obstacles to guide future research.

Prediction of Cognitive Impairment Risk among Older Adults: A Machine Learning-Based Comparative Study and Model Development

INTRODUCTION

The prevalence of cognitive impairment and dementia in the older population is increasing, and thereby, early detection of cognitive decline is essential for effective intervention.

METHODS

This study included 2,288 participants with normal cognitive function from the Ma'anshan Healthy Aging Cohort Study. Forty-two potential predictors, including demographic characteristics, chronic diseases, lifestyle factors, anthropometric indices, physical function, and baseline cognitive function, were selected based on clinical importance and previous research. The dataset was partitioned into training, validation, and test sets in a proportion of 60% for training, 20% for validation, and 20% for testing, respectively. Recursive feature elimination was used for feature selection, followed by six machine learning algorithms that were employed for model development. The performance of the models was evaluated using area under the curve (AUC), specificity, sensitivity, and accuracy. Moreover, SHapley Additive exPlanations (SHAP) was conducted to access the interpretability of the final selected model and to gain insights into the impact of features on the prediction outcomes. SHAP force plots were established to vividly show the application of the prediction model at the individual level.

RESULTS

The final predictive model based on the Naive Bayes algorithm achieved an AUC of 0.820 (95% CI, 0.773-0.887) on the test set, outperforming other algorithms. The top ten influential features in the model included baseline Mini-Mental State Examination (MMSE), education, self-reported economic status, collective or social activities, Pittsburgh sleep quality index (PSQI), body mass index, systolic blood pressure, diastolic blood pressure, instrumental activities of daily living, and age. The model demonstrated the potential to identify individuals at a higher risk of cognitive impairment within 3 years from older adults.

CONCLUSION

The predictive model developed in this study contributes to the early detection of cognitive impairment in older adults by primary healthcare staff in community settings.

AI-based differential diagnosis of dementia etiologies on multimodal data

Differential diagnosis of dementia remains a challenge in neurology due to symptom overlap across etiologies, yet it is crucial for formulating early, personalized management strategies. Here, we present an AI model that harnesses a broad array of data, including demographics, individual and family medical history, medication use, neuropsychological assessments, functional evaluations, and multimodal neuroimaging, to identify the etiologies contributing to dementia in individuals. The study, drawing on 51,269 participants across 9 independent, geographically diverse datasets, facilitated the identification of 10 distinct dementia etiologies. It aligns diagnoses with similar management strategies, ensuring robust predictions even with incomplete data. Our model achieved a micro-averaged area under the receiver operating characteristic curve (AUROC) of 0.94 in classifying individuals with normal cognition, mild cognitive impairment and dementia. Also, the micro-averaged AUROC was 0.96 in differentiating the dementia etiologies. Our model demonstrated proficiency in addressing mixed dementia cases, with a mean AUROC of 0.78 for two co-occurring pathologies. In a randomly selected subset of 100 cases, the AUROC of neurologist assessments augmented by our AI model exceeded neurologist-only evaluations by 26.25%. Furthermore, our model predictions aligned with biomarker evidence and its associations with different proteinopathies were substantiated through postmortem findings. Our framework has the potential to be integrated as a screening tool for dementia in various clinical settings and drug trials, with promising implications for person-level management.

Prediction of positive pulmonary nodules based on machine learning algorithm combined with central carbon metabolism data

BACKGROUND

Lung cancer causes a huge disease burden, and early detection of positive pulmonary nodules (PPNs) as an early sign of lung cancer is extremely important for effective intervention. It is necessary to develop PPNs risk recognizer based on machine learning algorithm combined with central carbon metabolomics.

METHODS

The study included 2248 participants at high risk for lung cancer from the Ma'anshan Community Lung Cancer Screening cohort. The Least Absolute Shrinkage and Selection Operator (LASSO) was used to screen 18 central carbon-related metabolites in plasma, recursive feature elimination (RFE) was used to select all 42 features, followed by five machine learning algorithms for model development. The performance of the model was evaluated using area under the receiver operator characteristic curve (AUC), accuracy, precision, recall, and F1 scores. In addition, SHapley Additive exPlanations (SHAP) was performed to assess the interpretability of the final selected model and to gain insight into the impact of features on the predicted results.

RESULTS

Finally, the two prediction models based on the random forest (RF) algorithm performed best, with AUC values of 0.87 and 0.83, respectively, better than other models. We found that homogentisic acid, fumaric acid, maleic acid, hippuric acid, gluconic acid, and succinic acid played a significant role in both PPNs prediction model and NPNs vs PPNs model, while 2-oxadipic acid only played a role in the former model and phosphopyruvate only played a role in the NPNs vs PPNs model. This model demonstrates the potential of central carbon metabolism for PPNs risk prediction and identification.

CONCLUSION

We developed a series of predictive models for PPNs, which can help in the early detection of PPNs and thus reduce the risk of lung cancer.

Artificial intelligence for dementia research methods optimization

Artificial intelligence (AI) and machine learning (ML) approaches are increasingly being used in dementia research. However, several methodological challenges exist that may limit the insights we can obtain from high-dimensional data and our ability to translate these findings into improved patient outcomes. To improve reproducibility and replicability, researchers should make their well-documented code and modeling pipelines openly available. Data should also be shared where appropriate. To enhance the acceptability of models and AI-enabled systems to users, researchers should prioritize interpretable methods that provide insights into how decisions are generated. Models should be developed using multiple, diverse datasets to improve robustness, generalizability, and reduce potentially harmful bias. To improve clarity and reproducibility, researchers should adhere to reporting guidelines that are co-produced with multiple stakeholders. If these methodological challenges are overcome, AI and ML hold enormous promise for changing the landscape of dementia research and care. HIGHLIGHTS: Machine learning (ML) can improve diagnosis, prevention, and management of dementia. Inadequate reporting of ML procedures affects reproduction/replication of results. ML models built on unrepresentative datasets do not generalize to new datasets. Obligatory metrics for certain model structures and use cases have not been defined. Interpretability and trust in ML predictions are barriers to clinical translation.

Artificial intelligence for dementia prevention

INTRODUCTION

A wide range of modifiable risk factors for dementia have been identified. Considerable debate remains about these risk factors, possible interactions between them or with genetic risk, and causality, and how they can help in clinical trial recruitment and drug development. Artificial intelligence (AI) and machine learning (ML) may refine understanding.

METHODS

ML approaches are being developed in dementia prevention. We discuss exemplar uses and evaluate the current applications and limitations in the dementia prevention field.

RESULTS

Risk-profiling tools may help identify high-risk populations for clinical trials; however, their performance needs improvement. New risk-profiling and trial-recruitment tools underpinned by ML models may be effective in reducing costs and improving future trials. ML can inform drug-repurposing efforts and prioritization of disease-modifying therapeutics.

DISCUSSION

ML is not yet widely used but has considerable potential to enhance precision in dementia prevention.

HIGHLIGHTS

Artificial intelligence (AI) is not widely used in the dementia prevention field. Risk-profiling tools are not used in clinical practice. Causal insights are needed to understand risk factors over the lifespan. AI will help personalize risk-management tools for dementia prevention. AI could target specific patient groups that will benefit most for clinical trials.

Data structuring may prevent ambiguity and improve personalized medical prognosis

Topics expected to influence personalized medicine (PM), where medical decisions, practices, and treatments are tailored to the individual patient, are reviewed. Lack of discrimination due to different biological conditions that express similar values of numerical variables (ambiguity) is regarded to be a major potential barrier for PM. This material explores possible causes and sources of ambiguity and offers suggestions for mitigating the impacts of uncertainties. Three causes of ambiguity are identified: (1) delayed adoption of innovations, (2) inadequate emphases, and (3) inadequate processes used when new medical practices are developed and validated. One example of the first problem is the relative lack of medical research on "compositional data" -the type that characterizes leukocyte data. This omission results in erroneous use of data abundantly utilized in medicine, such as the blood cell differential. Emphasis on data output ‒not biomedical interpretation that facilitates the use of clinical data‒ exemplifies the second type of problems. Reliance on tools generated in other fields (but not validated within biomedical contexts) describes the last limitation. Because reductionism is associated with these problems, non-reductionist alternatives are reviewed as potential remedies. Data structuring (converting data into information) is considered a key element that may promote PM. To illustrate a process that includes data-information-knowledge and decision-making, previously published data on COVID-19 are utilized. It is suggested that ambiguity may be prevented or ameliorated. Provided that validations are grounded on biomedical knowledge, approaches that describe certain criteria - such as non-overlapping data intervals of patients that experience different outcomes, immunologically interpretable data, and distinct graphic patterns - can inform, at personalized bases, earlier and/or with fewer observations.

Metallomic profiling and natural copper isotopic signatures of childhood autism in serum and red blood cells

Excessive exposure to metals directly threatens human health, including neurodeve lopment. Autism spectrum disorder (ASD) is a neurodevelopmental disorder, leaving great harms to children themselves, their families, and even society. In view of this, it is critical to develop reliable biomarkers for ASD in early childhood. Here we used inductively coupled plasma mass spectrometry (ICP-MS) to identify the abnormalities in ASD-associated metal elements in children blood. Multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS) was applied to detect isotopic differences in copper (Cu) for further assessment on account of its core role in the brain. We also developed a machine learning classification method for unknown samples based on a support vector machine (SVM) algorithm. The results indicated significant differences in the blood metallome (chromium (Cr), manganese (Mn), cobalt (Co), magnesium (Mg), and arsenic (As)) between cases and controls, and a significantly lower Zn/Cu ratio was observed in the ASD cases. Interestingly, we found a strong association of serum copper isotopic composition (δ⁶⁵Cu) with autistic serum. SVM was successfully applied to discriminate cases and controls based on the two-dimensional Cu signatures (Cu concentration and δ⁶⁵Cu) with a high accuracy (94.4%). Overall, our findings revealed a new biomarker for potential early diagnosis and screening of ASD, and the significant alterations in the blood metallome also helped to understand the potential pathogenesis of ASD in terms of metallomics.

Evaluation of artificial intelligence techniques in disease diagnosis and prediction

A broad range of medical diagnoses is based on analyzing disease images obtained through high-tech digital devices. The application of artificial intelligence (AI) in the assessment of medical images has led to accurate evaluations being performed automatically, which in turn has reduced the workload of physicians, decreased errors and times in diagnosis, and improved performance in the prediction and detection of various diseases. AI techniques based on medical image processing are an essential area of research that uses advanced computer algorithms for prediction, diagnosis, and treatment planning, leading to a remarkable impact on decision-making procedures. Machine Learning (ML) and Deep Learning (DL) as advanced AI techniques are two main subfields applied in the healthcare system to diagnose diseases, discover medication, and identify patient risk factors. The advancement of electronic medical records and big data technologies in recent years has accompanied the success of ML and DL algorithms. ML includes neural networks and fuzzy logic algorithms with various applications in automating forecasting and diagnosis processes. DL algorithm is an ML technique that does not rely on expert feature extraction, unlike classical neural network algorithms. DL algorithms with high-performance calculations give promising results in medical image analysis, such as fusion, segmentation, recording, and classification. Support Vector Machine (SVM) as an ML method and Convolutional Neural Network (CNN) as a DL method is usually the most widely used techniques for analyzing and diagnosing diseases. This review study aims to cover recent AI techniques in diagnosing and predicting numerous diseases such as cancers, heart, lung, skin, genetic, and neural disorders, which perform more precisely compared to specialists without human error. Also, AI's existing challenges and limitations in the medical area are discussed and highlighted.

Whole-brain dynamical modelling for classification of Parkinson's disease

Simulated whole-brain connectomes demonstrate enhanced inter-individual variability depending on the data processing and modelling approach. By considering the human brain connectome as an individualized attribute, we investigate how empirical and simulated whole-brain connectome-derived features can be utilized to classify patients with Parkinson's disease against healthy controls in light of varying data processing and model validation. To this end, we applied simulated blood oxygenation level-dependent signals derived by a whole-brain dynamical model simulating electrical signals of neuronal populations to reveal differences between patients and controls. In addition to the widely used model validation via fitting the dynamical model to empirical neuroimaging data, we invented a model validation against behavioural data, such as subject classes, which we refer to as behavioural model fitting and show that it can be beneficial for Parkinsonian patient classification. Furthermore, the results of machine learning reported in this study also demonstrated that the performance of the patient classification can be improved when the empirical data are complemented by the simulation results. We also showed that the temporal filtering of blood oxygenation level-dependent signals influences the prediction results, where filtering in the low-frequency band is advisable for Parkinsonian patient classification. In addition, composing the feature space of empirical and simulated data from multiple brain parcellation schemes provided complementary features that improved prediction performance. Based on our findings, we suggest that combining the simulation results with empirical data is effective for inter-individual research and its clinical application.

Graph-generative neural network for EEG-based epileptic seizure detection via discovery of dynamic brain functional connectivity

Dynamic complexity in brain functional connectivity has hindered the effective use of signal processing or machine learning methods to diagnose neurological disorders such as epilepsy. This paper proposed a new graph-generative neural network (GGN) model for the dynamic discovery of brain functional connectivity via deep analysis of scalp electroencephalogram (EEG) signals recorded from various regions of a patient's scalp. Brain functional connectivity graphs are generated for the extraction of spatial-temporal resolution of various onset epilepsy seizure patterns. Our supervised GGN model was substantiated by seizure detection and classification experiments. We train the GGN model using a clinically proven dataset of over 3047 epileptic seizure cases. The GGN model achieved a 91% accuracy in classifying seven types of epileptic seizure attacks, which outperformed the 65%, 74%, and 82% accuracy in using the convolutional neural network (CNN), graph neural networks (GNN), and transformer models, respectively. We present the GGN model architecture and operational steps to assist neuroscientists or brain specialists in using dynamic functional connectivity information to detect neurological disorders. Furthermore, we suggest to merge our spatial-temporal graph generator design in upgrading the conventional CNN and GNN models with dynamic convolutional kernels for accuracy enhancement.

Automated differential diagnosis of dementia syndromes using FDG PET and machine learning

BACKGROUND

Metabolic brain imaging with 2-[¹⁸F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET) is a supportive diagnostic and differential diagnostic tool for neurodegenerative dementias. In the clinic, scans are usually visually interpreted. However, computer-aided approaches can improve diagnostic accuracy. We aimed to build two machine learning classifiers, based on two sets of FDG PET-derived features, for differential diagnosis of common dementia syndromes.

METHODS

We analyzed FDG PET scans from three dementia cohorts [63 dementia due to Alzheimer's disease (AD), 79 dementia with Lewy bodies (DLB) and 23 frontotemporal dementia (FTD)], and 41 normal controls (NCs). Patients' clinical diagnosis at follow-up (25 ± 20 months after scanning) or cerebrospinal fluid biomarkers for Alzheimer's disease was considered a gold standard. FDG PET scans were first visually evaluated. Scans were pre-processed, and two sets of features extracted: (1) the expressions of previously identified metabolic brain patterns, and (2) the mean uptake value in 95 regions of interest (ROIs). Two multi-class support vector machine (SVM) classifiers were tested and their diagnostic performance assessed and compared to visual reading. Class-specific regional feature importance was assessed with Shapley Additive Explanations.

RESULTS

Pattern- and ROI-based classifier achieved higher overall accuracy than expert readers (78% and 80% respectively, vs. 71%). Both SVM classifiers performed similarly to one another and to expert readers in AD (F1 = 0.74, 0.78, and 0.78) and DLB (F1 = 0.81, 0.81, and 0.78). SVM classifiers outperformed expert readers in FTD (F1 = 0.87, 0.83, and 0.63), but not in NC (F1 = 0.71, 0.75, and 0.92). Visualization of the SVM model showed bilateral temporal cortices and cerebellum to be the most important features for AD; occipital cortices, hippocampi and parahippocampi, amygdala, and middle temporal lobes for DLB; bilateral frontal cortices, middle and anterior cingulum for FTD; and bilateral angular gyri, pons, and vermis for NC.

CONCLUSION

Multi-class SVM classifiers based on the expression of characteristic metabolic brain patterns or ROI glucose uptake, performed better than experts in the differential diagnosis of common dementias using FDG PET scans. Experts performed better in the recognition of normal scans and a combined approach may yield optimal results in the clinical setting.

NigraNet: An automatic framework to assess nigral neuromelanin content in early Parkinson's disease using convolutional neural network

BACKGROUND

Parkinson's disease (PD) demonstrates neurodegenerative changes in the substantia nigra pars compacta (SNc) using neuromelanin-sensitive (NM)-MRI. As SNc manual segmentation is prone to substantial inter-individual variability across raters, development of a robust automatic segmentation framework is necessary to facilitate nigral neuromelanin quantification. Artificial intelligence (AI) is gaining traction in the neuroimaging community for automated brain region segmentation tasks using MRI.

OBJECTIVE

Developing and validating AI-based NigraNet, a fully automatic SNc segmentation framework allowing nigral neuromelanin quantification in patients with PD using NM-MRI.

METHODS

We prospectively included 199 participants comprising 144 early-stage idiopathic PD patients (disease duration = 1.5 ± 1.0 years) and 55 healthy volunteers (HV) scanned using a 3 Tesla MRI including whole brain T1-weighted anatomical imaging and NM-MRI. The regions of interest (ROI) were delineated in all participants automatically using NigraNet, a modified U-net, and compared to manual segmentations performed by two experienced raters. The SNc volumes (Vol), volumes corrected by total intracranial volume (Cvol), normalized signal intensity (NSI) and contrast-to-noise ratio (CNR) were computed. One-way GLM-ANCOVA was performed while adjusting for age and sex as covariates. Diagnostic performance measurement was assessed using the receiver operating characteristic (ROC) analysis. Inter and intra-observer variability were estimated using Dice similarity coefficient (DSC). The agreements between methods were tested using intraclass correlation coefficient (ICC) based on a mean-rating, two-way, mixed-effects model estimates for absolute agreement. Cronbach's alpha and Bland-Altman plots were estimated to assess inter-method consistency.

RESULTS

Using both methods, Vol, Cvol, NSI and CNR measurements differed between PD and HV with an effect of sex for Cvol and CNR. ICC values between the methods demonstrated optimal agreement for Cvol and CNR (ICC > 0.9) and high reproducibility (DSC: 0.80) was also obtained. The SNc measurements also showed good to excellent consistency values (Cronbach's alpha > 0.87). Bland-Altman plots of agreement demonstrated no association of SNc ROI measurement differences between the methods and ROI average measurements while confirming that 95 % of the data points were ranging between the limits of mean difference (d ± 1.96xSD). Percentage changes between PD and HV were -27.4 % and -17.7 % for Vol, -30.0 % and -22.2 % for Cvol, -15.8 % and -14.4 % for NSI, -17.1 % and -16.0 % for CNR for automatic and manual measurements respectively. Using automatic method, in the entire dataset, we obtained the areas under the ROC curve (AUC) of 0.83 for Vol, 0.85 for Cvol, 0.79 for NSI and 0.77 for CNR whereas in the training dataset of 0.96 for Vol, 0.95 for Cvol, 0.85 for NSI and 0.85 for CNR. Disease duration correlated negatively with NSI of the patients for both the automatic and manual measurements.

CONCLUSIONS

We presented an AI-based NigraNet framework that utilizes a small MRI training dataset to fully automatize the SNc segmentation procedure with an increased precision and more reproducible results. Considering the consistency, accuracy and speed of our approach, this study could be a crucial step towards the implementation of a time-saving non-rater dependent fully automatic method for studying neuromelanin changes in clinical settings and large-scale neuroimaging studies.

Ethical considerations for precision psychiatry: A roadmap for research and clinical practice

Precision psychiatry is an emerging field with transformative opportunities for mental health. However, the use of clinical prediction models carries unprecedented ethical challenges, which must be addressed before accessing the potential benefits of precision psychiatry. This critical review covers multidisciplinary areas, including psychiatry, ethics, statistics and machine-learning, healthcare and academia, as well as input from people with lived experience of mental disorders, their family, and carers. We aimed to identify core ethical considerations for precision psychiatry and mitigate concerns by designing a roadmap for research and clinical practice. We identified priorities: learning from somatic medicine; identifying precision psychiatry use cases; enhancing transparency and generalizability; fostering implementation; promoting mental health literacy; communicating risk estimates; data protection and privacy; and fostering the equitable distribution of mental health care. We hope this blueprint will advance research and practice and enable people with mental health problems to benefit from precision psychiatry.

Machine Learning Techniques for the Diagnosis of Schizophrenia Based on Event-Related Potentials

Antecedent

The event-related potential (ERP) components P300 and mismatch negativity (MMN) have been linked to cognitive deficits in patients with schizophrenia. The diagnosis of schizophrenia could be improved by applying machine learning procedures to these objective neurophysiological biomarkers. Several studies have attempted to achieve this goal, but no study has examined Multiple Kernel Learning (MKL) classifiers. This algorithm finds optimally a combination of kernel functions, integrating them in a meaningful manner, and thus could improve diagnosis.

Objective

This study aimed to examine the efficacy of the MKL classifier and the Boruta feature selection method for schizophrenia patients (SZ) and healthy controls (HC) single-subject classification.

Methods

A cohort of 54 SZ and 54 HC participants were studied. Three sets of features related to ERP signals were calculated as follows: peak related features, peak to peak related features, and signal related features. The Boruta algorithm was used to evaluate the impact of feature selection on classification performance. An MKL algorithm was applied to address schizophrenia detection.

Results

A classification accuracy of 83% using the whole dataset, and 86% after applying Boruta feature selection was obtained. The variables that contributed most to the classification were mainly related to the latency and amplitude of the auditory P300 paradigm.

Conclusion

This study showed that MKL can be useful in distinguishing between schizophrenic patients and controls when using ERP measures. Moreover, the use of the Boruta algorithm provides an improvement in classification accuracy and computational cost.

Diagnosis of middle cerebral artery stenosis using the transcranial Doppler images based on convolutional neural network

BACKGROUND

The purpose of this study was to explore the diagnostic value of convolutional neural networks (CNNs) in middle cerebral artery (MCA) stenosis by analyzing the transcranial Doppler (TCD) images.

METHODS

Overall 278 patients who underwent cerebral vascular TCD and cerebral angiography were enrolled and classified into stenosis and non-stenosis groups based on cerebral angiography findings. Manual measurements were performed on TCD images. The patients were divided into a training set and a test set, and the CNNs architecture was used to classify TCD images. The diagnostic accuracies of manual measurements, CNNs, and TCD parameters for MCA stenosis were calculated and compared.

RESULTS

Overall, 203 patients without stenosis and 75 patients with stenosis were evaluated. The sensitivity, specificity, and area under the curve (AUC) for manual measurements of MCA stenosis were 0.80, 0.83, and 0.81, respectively. After 24 iterations of the running model in the training set, the sensitivity, specificity, and AUC of the CNNs in the test set were 0.84, 0.86, and 0.80, respectively. The diagnostic value of CNNs differed minimally from that of manual measurements. Two parameters of TCD, peak systolic velocity and mean flow velocity, were higher in patients with stenosis than in those without stenosis; however, their diagnostic values were significantly lower than those of CNNs (P < 0.05).

CONCLUSIONS

The diagnostic value of CNNs for MCA stenosis based on TCD images paralleled that of manual measurements. CNNs could be used as an auxiliary diagnostic tool to improve the diagnosis of MCA stenosis.

A Survey on EEG Signal Processing Techniques and Machine Learning: Applications to the Neurofeedback of Autobiographical Memory Deficits in Schizophrenia

In this paper, a general overview regarding neural recording, classical signal processing techniques and machine learning classification algorithms applied to monitor brain activity is presented. Currently, several approaches classified as electrical, magnetic, neuroimaging recordings and brain stimulations are available to obtain neural activity of the human brain. Among them, non-invasive methods like electroencephalography (EEG) are commonly employed, as they can provide a high degree of temporal resolution (on the order of milliseconds) and acceptable space resolution. In addition, it is simple, quick, and does not create any physical harm or stress to patients. Concerning signal processing, once the neural signals are acquired, different procedures can be applied for feature extraction. In particular, brain signals are normally processed in time, frequency, and/or space domains. The features extracted are then used for signal classification depending on its characteristics such us the mean, variance or band power. The role of machine learning in this regard has become of key importance during the last years due to its high capacity to analyze complex amounts of data. The algorithms employed are generally classified in supervised, unsupervised and reinforcement techniques. A deep review of the most used machine learning algorithms and the advantages/drawbacks of most used methods is presented. Finally, a study of these procedures utilized in a very specific and novel research field of electroencephalography, i.e., autobiographical memory deficits in schizophrenia, is outlined.

Performance of Machine Learning Algorithms for Predicting Progression to Dementia in Memory Clinic Patients

IMPORTANCE

Machine learning algorithms could be used as the basis for clinical decision-making aids to enhance clinical practice.

OBJECTIVE

To assess the ability of machine learning algorithms to predict dementia incidence within 2 years compared with existing models and determine the optimal analytic approach and number of variables required.

DESIGN, SETTING, AND PARTICIPANTS

This prognostic study used data from a prospective cohort of 15 307 participants without dementia at baseline to perform a secondary analysis of factors that could be used to predict dementia incidence. Participants attended National Alzheimer Coordinating Center memory clinics across the United States between 2005 and 2015. Analyses were conducted from March to May 2021.

EXPOSURES

258 variables spanning domains of dementia-related clinical measures and risk factors.

MAIN OUTCOMES AND MEASURES

The main outcome was incident all-cause dementia diagnosed within 2 years of baseline assessment.

RESULTS

In a sample of 15 307 participants (mean [SD] age, 72.3 [9.8] years; 9129 [60%] women and 6178 [40%] men) without dementia at baseline, 1568 (10%) received a diagnosis of dementia within 2 years of their initial assessment. Compared with 2 existing models for dementia risk prediction (ie, Cardiovascular Risk Factors, Aging, and Incidence of Dementia Risk Score, and the Brief Dementia Screening Indicator), machine learning algorithms were superior in predicting incident all-cause dementia within 2 years. The gradient-boosted trees algorithm had a mean (SD) overall accuracy of 92% (1%), sensitivity of 0.45 (0.05), specificity of 0.97 (0.01), and area under the curve of 0.92 (0.01) using all 258 variables. Analysis of variable importance showed that only 6 variables were required for machine learning algorithms to achieve an accuracy of 91% and area under the curve of at least 0.89. Machine learning algorithms also identified up to 84% of participants who received an initial dementia diagnosis that was subsequently reversed to mild cognitive impairment or cognitively unimpaired, suggesting possible misdiagnosis.

CONCLUSIONS AND RELEVANCE

These findings suggest that machine learning algorithms could accurately predict incident dementia within 2 years in patients receiving care at memory clinics using only 6 variables. These findings could be used to inform the development and validation of decision-making aids in memory clinics.

Automatic quality control of brain T1-weighted magnetic resonance images for a clinical data warehouse

Many studies on machine learning (ML) for computer-aided diagnosis have so far been mostly restricted to high-quality research data. Clinical data warehouses, gathering routine examinations from hospitals, offer great promises for training and validation of ML models in a realistic setting. However, the use of such clinical data warehouses requires quality control (QC) tools. Visual QC by experts is time-consuming and does not scale to large datasets. In this paper, we propose a convolutional neural network (CNN) for the automatic QC of 3D T1-weighted brain MRI for a large heterogeneous clinical data warehouse. To that purpose, we used the data warehouse of the hospitals of the Greater Paris area (Assistance Publique-Hôpitaux de Paris [AP-HP]). Specifically, the objectives were: 1) to identify images which are not proper T1-weighted brain MRIs; 2) to identify acquisitions for which gadolinium was injected; 3) to rate the overall image quality. We used 5000 images for training and validation and a separate set of 500 images for testing. In order to train/validate the CNN, the data were annotated by two trained raters according to a visual QC protocol that we specifically designed for application in the setting of a data warehouse. For objectives 1 and 2, our approach achieved excellent accuracy (balanced accuracy and F1-score >90%), similar to the human raters. For objective 3, the performance was good but substantially lower than that of human raters. Nevertheless, the automatic approach accurately identified (balanced accuracy and F1-score >80%) low quality images, which would typically need to be excluded. Overall, our approach shall be useful for exploiting hospital data warehouses in medical image computing.

Update on neuroimaging for categorization of Parkinson's disease and atypical parkinsonism

PURPOSE OF REVIEW

Differential diagnosis of Parkinsonism may be difficult. The objective of this review is to present the work of the last three years in the field of imaging for diagnostic categorization of parkinsonian syndromes focusing on progressive supranuclear palsy (PSP) and multiple system atrophy (MSA).

RECENT FINDINGS

Two main complementary approaches are being pursued. The first seeks to develop and validate manual qualitative or semi-quantitative imaging markers that can be easily used in clinical practice. The second is based on quantitative measurements of magnetic resonance imaging abnormalities integrated in a multimodal approach and in automatic categorization machine learning tools.

SUMMARY

These two complementary approaches obtained high diagnostic around 90% and above in the classical Richardson form of PSP and probable MSA. Future work will determine if these techniques can improve diagnosis in other PSP variants and early forms of the diseases when all clinical criteria are not fully met.

Deep Learning-Based Classification and Voxel-Based Visualization of Frontotemporal Dementia and Alzheimer's Disease

Frontotemporal dementia (FTD) and Alzheimer's disease (AD) have overlapping symptoms, and accurate differential diagnosis is important for targeted intervention and treatment. Previous studies suggest that the deep learning (DL) techniques have the potential to solve the differential diagnosis problem of FTD, AD and normal controls (NCs), but its performance is still unclear. In addition, existing DL-assisted diagnostic studies still rely on hypothesis-based expert-level preprocessing. On the one hand, it imposes high requirements on clinicians and data themselves; On the other hand, it hinders the backtracking of classification results to the original image data, resulting in the classification results cannot be interpreted intuitively. In the current study, a large cohort of 3D T1-weighted structural magnetic resonance imaging (MRI) volumes (n = 4,099) was collected from two publicly available databases, i.e., the ADNI and the NIFD. We trained a DL-based network directly based on raw T1 images to classify FTD, AD and corresponding NCs. And we evaluated the convergence speed, differential diagnosis ability, robustness and generalizability under nine scenarios. The proposed network yielded an accuracy of 91.83% based on the most common T1-weighted sequence [magnetization-prepared rapid acquisition with gradient echo (MPRAGE)]. The knowledge learned by the DL network through multiple classification tasks can also be used to solve subproblems, and the knowledge is generalizable and not limited to a specified dataset. Furthermore, we applied a gradient visualization algorithm based on guided backpropagation to calculate the contribution graph, which tells us intuitively why the DL-based networks make each decision. The regions making valuable contributions to FTD were more widespread in the right frontal white matter regions, while the left temporal, bilateral inferior frontal and parahippocampal regions were contributors to the classification of AD. Our results demonstrated that DL-based networks have the ability to solve the enigma of differential diagnosis of diseases without any hypothesis-based preprocessing. Moreover, they may mine the potential patterns that may be different from human clinicians, which may provide new insight into the understanding of FTD and AD.

Deep learning for brain disorders: from data processing to disease treatment

In order to reach precision medicine and improve patients' quality of life, machine learning is increasingly used in medicine. Brain disorders are often complex and heterogeneous, and several modalities such as demographic, clinical, imaging, genetics and environmental data have been studied to improve their understanding. Deep learning, a subpart of machine learning, provides complex algorithms that can learn from such various data. It has become state of the art in numerous fields, including computer vision and natural language processing, and is also growingly applied in medicine. In this article, we review the use of deep learning for brain disorders. More specifically, we identify the main applications, the concerned disorders and the types of architectures and data used. Finally, we provide guidelines to bridge the gap between research studies and clinical routine.

Home-Based Functional Electrical Stimulation of Human Permanent Denervated Muscles: A Narrative Review on Diagnostics, Managements, Results and Byproducts Revisited 2020

Spinal cord injury (SCI) produces muscle wasting that is especially severe after complete and permanent damage of lower motor neurons, as can occur in complete conus and cauda equina syndrome. Even in this worst-case scenario, mass and function of permanently denervated quadriceps muscle can be rescued by surface functional electrical stimulation using a purpose designed home-based rehabilitation strategy. Early diagnostics is a key factor in the long-term success of this management. Function of quadriceps muscle was quantitated by force measurements. Muscle gross cross-sections were evaluated by quantitative color computed tomography (CT) and muscle and skin biopsies by quantitative histology, electron microscopy, and immunohistochemistry. Two years of treatment that started earlier than 5 years from SCI produced: (a) an increase in cross-sectional area of stimulated muscles; (b) an increase in muscle fiber mean diameter; (c) improvements in ultrastructural organization; and (d) increased force output during electrical stimulation. Improvements are extended to hamstring muscles and skin. Indeed, the cushioning effect provided by recovered tissues is a major clinical benefit. It is our hope that new trials start soon, providing patients the benefits they need.