Brainomaly: Unsupervised Neurologic Disease Detection Utilizing Unannotated T1-weighted Brain MR Images.

Harnessing the power of deep neural networks in the medical imaging domain is challenging due to the difficulties in acquiring large annotated datasets, especially for rare diseases, which involve high costs, time, and effort for annotation. Unsupervised disease detection methods, such as anomaly detection, can significantly reduce human effort in these scenarios. While anomaly detection typically focuses on learning from images of healthy subjects only, real-world situations often present unannotated datasets with a mixture of healthy and diseased subjects. Recent studies have demonstrated that utilizing such unannotated images can improve unsupervised disease and anomaly detection. However, these methods do not utilize knowledge specific to registered neuroimages, resulting in a subpar performance in neurologic disease detection. To address this limitation, we propose Brainomaly, a GAN-based image-to-image translation method specifically designed for neurologic disease detection. Brainomaly not only offers tailored image-to-image translation suitable for neuroimages but also leverages unannotated mixed images to achieve superior neurologic disease detection. Additionally, we address the issue of model selection for inference without annotated samples by proposing a pseudo-AUC metric, further enhancing Brainomaly's detection performance. Extensive experiments and ablation studies demonstrate that Brainomaly outperforms existing state-of-the-art unsupervised disease and anomaly detection methods by significant margins in Alzheimer's disease detection using a publicly available dataset and headache detection using an institutional dataset. The code is available from https://github.com/mahfuzmohammad/Brainomaly.

Ordinal Classification with Distance Regularization for Robust Brain Age Prediction.

Age is one of the major known risk factors for Alzheimer's Disease (AD). Detecting AD early is crucial for effective treatment and preventing irreversible brain damage. Brain age, a measure derived from brain imaging reflecting structural changes due to aging, may have the potential to identify AD onset, assess disease risk, and plan targeted interventions. Deep learning-based regression techniques to predict brain age from magnetic resonance imaging (MRI) scans have shown great accuracy recently. However, these methods are subject to an inherent regression to the mean effect, which causes a systematic bias resulting in an overestimation of brain age in young subjects and underestimation in old subjects. This weakens the reliability of predicted brain age as a valid biomarker for downstream clinical applications. Here, we reformulate the brain age prediction task from regression to classification to address the issue of systematic bias. Recognizing the importance of preserving ordinal information from ages to understand aging trajectory and monitor aging longitudinally, we propose a novel ORdinal Distance Encoded Regularization (ORDER) loss that incorporates the order of age labels, enhancing the model's ability to capture age-related patterns. Extensive experiments and ablation studies demonstrate that this framework reduces systematic bias, outperforms state-of-art methods by statistically significant margins, and can better capture subtle differences between clinical groups in an independent AD dataset. Our implementation is publicly available at https://github.com/jaygshah/Robust-Brain-Age-Prediction.

Brain Structural and Functional Abnormalities Associated with Acute Post-Traumatic Headache: Iron Deposition and Functional Connectivity.

Background: The purpose of this study was to interrogate brain iron accumulation in participants with acute post-traumatic headache (PTH) due to mild traumatic brain injury (mTBI), and to determine if functional connectivity is affected in areas with iron accumulation. We aimed to examine the correlations between iron accumulation and headache frequency, post-concussion symptom severity, number of mTBIs and time since most recent TBI. Methods: Sixty participants with acute PTH and 60 age-matched healthy controls (HC) underwent 3T magnetic resonance imaging including quantitative T2* maps and resting-state functional connectivity imaging. Between group T2* differences were determined using T-tests (p < 0.005, cluster size threshold of 10 voxels). For regions with T2* differences, two analyses were conducted. First, the correlations with clinical variables including headache frequency, number of lifetime mTBIs, time since most recent mTBI, and Sport Concussion Assessment Tool (SCAT) symptom severity scale scores were investigated using linear regression. Second, the functional connectivity of these regions with the rest of the brain was examined (significance of p < 0.05 with family wise error correction for multiple comparisons). Results: The acute PTH group consisted of 60 participants (22 male, 38 female) with average age of 42 ± 14 years. The HC group consisted of 60 age-matched controls (17 male, 43 female, average age of 42 ± 13). PTH participants had lower T2* values compared to HC in the left posterior cingulate and the bilateral cuneus. Stronger functional connectivity was observed between bilateral cuneus and right cerebellar areas in PTH compared to HC. Within the PTH group, linear regression showed negative associations of T2* and SCAT symptom severity score in the left posterior cingulate (p = 0.05) and with headache frequency in the left cuneus (p = 0.04). Conclusions: Iron accumulation in posterior cingulate and cuneus was observed in those with acute PTH relative to HC; stronger functional connectivity was detected between the bilateral cuneus and the right cerebellum. The correlations of decreased T2* (suggesting higher iron content) with headache frequency and post mTBI symptom severity suggest that the iron accumulation that results from mTBI might reflect the severity of underlying mTBI pathophysiology and associate with post-mTBI symptom severity including PTH.

Improving newborn screening test performance for metachromatic leukodystrophy: Recommendation from a pre-pilot study that identified a late-infantile case for treatment.

Metachromatic leukodystrophy (MLD) is a devastating rare neurodegenerative disease. Typically, loss of motor and cognitive skills precedes early death. The disease is characterised by deficient lysosomal arylsulphatase A (ARSA) activity and an accumulation of undegraded sulphatide due to pathogenic variants in the ARSA gene. Atidarsagene autotemcel (arsa-cel), an ex vivo haematopoietic stem cell gene therapy was approved for use in the UK in 2021 to treat early-onset forms of pre- or early-symptomatic MLD. Optimal outcomes require early diagnosis, but in the absence of family history this is difficult to achieve without newborn screening (NBS). A pre-pilot MLD NBS study was conducted as a feasibility study in Manchester UK using a two-tiered screening test algorithm. Pre-established cutoff values (COV) for the first-tier C16:0 sulphatide (C16:0-S) and the second-tier ARSA tests were evaluated. Before the pre-pilot study, initial test validation using non­neonatal diagnostic bloodspots demonstrated ARSA pseudodeficiency status was associated with normal C16:0-S results for age (n = 43) and hence not expected to cause false positive results in this first-tier test. Instability of ARSA in bloodspot required transfer of NBS bloodspots from ambient temperature to -20°C storage within 7-8 days after heel prick, the earliest possible in this UK pre-pilot study. Eleven of 3687 de-identified NBS samples in the pre-pilot were positive for C16:0-S based on the pre-established COV of ≥170 nmol/l or ≥ 1.8 multiples of median (MoM). All 11 samples were subsequently tested negative determined by the ARSA COV of <20% mean of negative controls. However, two of 20 NBS samples from MLD patients would be missed by this C16:0-S COV. A further suspected false negative case that displayed 4% mean ARSA activity by single ARSA analysis for the initial test validation was confirmed by genotyping of this NBS bloodspot, a severe late infantile MLD phenotype was predicted. This led to urgent assessment of this child by authority approval and timely commencement of arsa-cel gene therapy at 11 months old. Secondary C16:0-S analysis of this NBS bloodspot was 150 nmol/l or 1.67 MoM. This was the lowest result reported thus far, a new COV of 1.65 MoM is recommended for future pilot studies. Furthermore, preliminary data of this study showed C16:1-OH sulphatide is more specific for MLD than C16:0-S. In conclusion, this pre-pilot study adds to the international evidence that recommends newborn screening for MLD, making it possible for patients to benefit fully from treatment through early diagnosis.

Higher precision, first tier newborn screening for metachromatic leukodystrophy using 16:1-OH-sulfatide.

Newborn screening (NBS) for metachromatic leukodystrophy (MLD) is based on first-tier measurement of sulfatides in dried blood spots (DBS) followed by second-tier measurement of arylsulfatase A in the same DBS. This approach is very precise with 0-1 false positives per ∼30,000 newborns tested. Recent data reported here shows that the sulfatide molecular species with an α-hydroxyl, 16­carbon, mono-unsaturated fatty acyl group (16:1-OH-sulfatide) is superior to the original biomarker 16:0-sulfatide in reducing the number of first-tier false positives. This result is consistent across 4 MLD NBS centers. By measuring 16:1-OH-sulfatide alone or together with 16:0-sulfatide, the estimated false positive rate is 0.048% and is reduced essentially to zero with second-tier arylsulfatase A activity assay. The false negative rate is predicted to be extremely low based on the demonstration that 40 out of 40 newborn DBS from clinically-confirmed MLD patients are detected with these methods. The work shows that NBS for MLD is extremely precise and ready for deployment. Furthermore, it can be multiplexed with several other inborn errors of metabolism already tested in NBS centers worldwide.

Retrospective Review of Positive Newborn Screening Results for Isovaleric Acidemia and Development of a Strategy to Improve the Efficacy of Newborn Screening in the UK.

Since the UK commenced newborn screening for isovaleric acidemia in 2015, changes in prescribing have increased the incidence of false positive (FP) results due to pivaloylcarnitine. A review of screening results between 2015 and 2022 identified 24 true positive (TP) and 84 FP cases, with pivalate interference confirmed in 76/84. Initial C5 carnitine (C5C) did not discriminate between FP and TP with median (range) C5C of 2.9 (2.0-9.6) and 4.0 (1.8->70) µmol/L, respectively, and neither did Precision Newborn Screening via Collaborative Laboratory Integrated Reports (CLIR), which identified only 1/47 FP cases. However, among the TP cases, disease severity showed a correlation with initial C5C in 'asymptomatic' individuals (n = 17), demonstrating a median (range) C5C of 3.0 (1.8-7.1) whilst 'clinically affected' patients (n = 7), showed a median (range) C5C of 13.9 (7.7-70) µmol/L. These findings allowed the introduction of dual cut-off values into the screening algorithm to reduce the incidence of FPs, with initial C5C results ≥ 5 µmol/L triggering urgent referral, and those >2.0 and <5.0 µmol/L prompting second-tier C5-isobar testing. This will avoid delayed referral in babies at particular risk whilst reducing the FP rate for the remainder.

Correlation enhanced distribution adaptation for prediction of fall risk.

With technological advancements in diagnostic imaging, smart sensing, and wearables, a multitude of heterogeneous sources or modalities are available to proactively monitor the health of the elderly. Due to the increasing risks of falls among older adults, an early diagnosis tool is crucial to prevent future falls. However, during the early stage of diagnosis, there is often limited or no labeled data (expert-confirmed diagnostic information) available in the target domain (new cohort) to determine the proper treatment for older adults. Instead, there are multiple related but non-identical domain data with labels from the existing cohort or different institutions. Integrating different data sources with labeled and unlabeled samples to predict a patient's condition poses a significant challenge. Traditional machine learning models assume that data for new patients follow a similar distribution. If the data does not satisfy this assumption, the trained models do not achieve the expected accuracy, leading to potential misdiagnosing risks. To address this issue, we utilize domain adaptation (DA) techniques, which employ labeled data from one or more related source domains. These DA techniques promise to tackle discrepancies in multiple data sources and achieve a robust diagnosis for new patients. In our research, we have developed an unsupervised DA model to align two domains by creating a domain-invariant feature representation. Subsequently, we have built a robust fall-risk prediction model based on these new feature representations. The results from simulation studies and real-world applications demonstrate that our proposed approach outperforms existing models.

Harmonizing florbetapir and PiB PET measurements of cortical Aß plaque burden using multiple regions-of-interest and machine learning techniques: An alternative to the Centiloid approach.

INTRODUCTION: Machine learning (ML) can optimize amyloid (Aß) comparability among positron emission tomography (PET) radiotracers. Using multi-regional florbetapir (FBP) measures and ML, we report better Pittsburgh compound-B (PiB)/FBP harmonization of mean-cortical Aß (mcAß) than Centiloid. METHODS: PiB-FBP pairs from 92 subjects in www.oasis-brains.org and 46 in www.gaain.org/centiloid-project were used as the training/testing sets. FreeSurfer-extracted FBP multi-regional Aß and actual PiB mcAß in the training set were used to train ML models generating synthetic PiB mcAß. The correlation coefficient (R) between the synthetic/actual PiB mcAß in the testing set was assessed. RESULTS: In the testing set, the synthetic/actual PiB mcAß correlation R = 0.985 (R2  = 0.970) using artificial neural network was significantly higher (p ≤ 6.6e-4) than the FBP/PiB correlation R = 0.927 (R2  = 0.860), improving total variance percentage (R2 ) from 86% to 97%. Other ML models such as partial least square, ensemble, and relevance vector regressions also improved R (p = 9.677e-05 /0.045/0.0017). DISCUSSION: ML improved mcAß comparability. Additional studies are needed for the generalizability to other amyloid tracers, and to tau PET. Highlights Centiloid is a calibration of the amyloid scale, not harmonization. Centiloid unifies the amyloid scale without improving inter-tracer association (R2 ). Machine learning (ML) can harmonize the amyloid scale by improving R2 . ML harmonization maps multi-regional florbetapir SUVRs to PiB mean-cortical SUVR. Artificial neural network ML increases Centiloid R2 from 86% to 97%.

BlobCUT: A Contrastive Learning Method to Support Small Blob Detection in Medical Imaging.

Medical imaging-based biomarkers derived from small objects (e.g., cell nuclei) play a crucial role in medical applications. However, detecting and segmenting small objects (a.k.a. blobs) remains a challenging task. In this research, we propose a novel 3D small blob detector called BlobCUT. BlobCUT is an unpaired image-to-image (I2I) translation model that falls under the Contrastive Unpaired Translation paradigm. It employs a blob synthesis module to generate synthetic 3D blobs with corresponding masks. This is incorporated into the iterative model training as the ground truth. The I2I translation process is designed with two constraints: (1) a convexity consistency constraint that relies on Hessian analysis to preserve the geometric properties and (2) an intensity distribution consistency constraint based on Kullback-Leibler divergence to preserve the intensity distribution of blobs. BlobCUT learns the inherent noise distribution from the target noisy blob images and performs image translation from the noisy domain to the clean domain, effectively functioning as a denoising process to support blob identification. To validate the performance of BlobCUT, we evaluate it on a 3D simulated dataset of blobs and a 3D MRI dataset of mouse kidneys. We conduct a comparative analysis involving six state-of-the-art methods. Our findings reveal that BlobCUT exhibits superior performance and training efficiency, utilizing only 56.6% of the training time required by the state-of-the-art BlobDetGAN. This underscores the effectiveness of BlobCUT in accurately segmenting small blobs while achieving notable gains in training efficiency.

Self-Supervised Contrastive Learning to Predict the Progression of Alzheimer's Disease with 3D Amyloid-PET.

Early diagnosis of Alzheimer's disease (AD) is an important task that facilitates the development of treatment and prevention strategies, and may potentially improve patient outcomes. Neuroimaging has shown great promise, including the amyloid-PET, which measures the accumulation of amyloid plaques in the brain-a hallmark of AD. It is desirable to train end-to-end deep learning models to predict the progression of AD for individuals at early stages based on 3D amyloid-PET. However, commonly used models are trained in a fully supervised learning manner, and they are inevitably biased toward the given label information. To this end, we propose a selfsupervised contrastive learning method to accurately predict the conversion to AD for individuals with mild cognitive impairment (MCI) with 3D amyloid-PET. The proposed method, SMoCo, uses both labeled and unlabeled data to capture general semantic representations underlying the images. As the downstream task is given as classification of converters vs. non-converters, unlike the general self-supervised learning problem that aims to generate task-agnostic representations, SMoCo additionally utilizes the label information in the pre-training. To demonstrate the performance of our method, we conducted experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. The results confirmed that the proposed method is capable of providing appropriate data representations, resulting in accurate classification. SMoCo showed the best classification performance over the existing methods, with AUROC = 85.17%, accuracy = 81.09%, sensitivity = 77.39%, and specificity = 82.17%. While SSL has demonstrated great success in other application domains of computer vision, this study provided the initial investigation of using a proposed self-supervised contrastive learning model, SMoCo, to effectively predict MCI conversion to AD based on 3D amyloid-PET.

A Mutual Knowledge Distillation-Empowered AI Framework for Early Detection of Alzheimer's Disease Using Incomplete Multi-Modal Images.

Early detection of Alzheimer's Disease (AD) is crucial to ensure timely interventions and optimize treatment outcomes for patients. While integrating multi-modal neuroimages, such as MRI and PET, has shown great promise, limited research has been done to effectively handle incomplete multi-modal image datasets in the integration. To this end, we propose a deep learning-based framework that employs Mutual Knowledge Distillation (MKD) to jointly model different sub-cohorts based on their respective available image modalities. In MKD, the model with more modalities (e.g., MRI and PET) is considered a teacher while the model with fewer modalities (e.g., only MRI) is considered a student. Our proposed MKD framework includes three key components: First, we design a teacher model that is student-oriented, namely the Student-oriented Multi-modal Teacher (SMT), through multi-modal information disentanglement. Second, we train the student model by not only minimizing its classification errors but also learning from the SMT teacher. Third, we update the teacher model by transfer learning from the student's feature extractor because the student model is trained with more samples. Evaluations on Alzheimer's Disease Neuroimaging Initiative (ADNI) datasets highlight the effectiveness of our method. Our work demonstrates the potential of using AI for addressing the challenges of incomplete multi-modal neuroimage datasets, opening new avenues for advancing early AD detection and treatment strategies.

Corrections to "STRIDE: Systematic Radar Intelligence Analysis for ADRD Risk Evaluation with Gait Signature Simulation and Deep Learning".

[This corrects the article PMC10399976.].

STRIDE: Systematic Radar Intelligence Analysis for ADRD Risk Evaluation with Gait Signature Simulation and Deep Learning.

Abnormal gait is a significant non-cognitive biomarker for Alzheimer's disease (AD) and AD-related dementia (ADRD). Micro-Doppler radar, a non-wearable technology, can capture human gait movements for potential early ADRD risk assessment. In this research, we propose to design STRIDE integrating micro-Doppler radar sensors with advanced artificial intelligence (AI) technologies. STRIDE embeds a new deep learning (DL) classification framework. As a proof of concept, we develop a "digital-twin" of STRIDE, consisting of a human walking simulation model and a micro-Doppler radar simulation model, to generate a gait signature dataset. Taking established human walking parameters, the walking model simulates individuals with ADRD under various conditions. The radar model based on electromagnetic scattering and the Doppler frequency shift model is employed to generate micro-Doppler signatures from different moving body parts (e.g., foot, limb, joint, torso, shoulder, etc.). A band-dependent DL framework is developed to predict ADRD risks. The experimental results demonstrate the effectiveness and feasibility of STRIDE for evaluating ADRD risk.

Harmonizing Healthy Cohorts to Support Multicenter Studies on Migraine Classification using Brain MRI Data.

Multicenter and multi-scanner imaging studies might be needed to provide sample sizes large enough for developing accurate predictive models. However, multicenter studies, which likely include confounding factors due to subtle differences in research participant characteristics, MRI scanners, and imaging acquisition protocols, might not yield generalizable machine learning models, that is, models developed using one dataset may not be applicable to a different dataset. The generalizability of classification models is key for multi-scanner and multicenter studies, and for providing reproducible results. This study developed a data harmonization strategy to identify healthy controls with similar (homogenous) characteristics from multicenter studies to validate the generalization of machine-learning techniques for classifying individual migraine patients and healthy controls using brain MRI data. The Maximum Mean Discrepancy (MMD) was used to compare the two datasets represented in Geodesic Flow Kernel (GFK) space, capturing the data variabilities for identifying a "healthy core". A set of homogeneous healthy controls can assist in overcoming some of the unwanted heterogeneity and allow for the development of classification models that have high accuracy when applied to new datasets. Extensive experimental results show the utilization of a healthy core. One dataset consists of 120 individuals (66 with migraine and 54 healthy controls) and another dataset consists of 76 (34 with migraine and 42 healthy controls) individuals. A homogeneous dataset derived from a cohort of healthy controls improves the performance of classification models by about 25% accuracy improvements for both episodic and chronic migraineurs.

Use of novel structural features to identify urinary biomarkers during acute kidney injury that predict progression to chronic kidney disease.

BACKGROUND: A significant barrier to biomarker development in the field of acute kidney injury (AKI) is the use of kidney function to identify candidates. Progress in imaging technology makes it possible to detect early structural changes prior to a decline in kidney function. Early identification of those who will advance to chronic kidney disease (CKD) would allow for the initiation of interventions to halt progression. The goal of this study was to use a structural phenotype defined by magnetic resonance imaging and histology to advance biomarker discovery during the transition from AKI to CKD. METHODS: Urine was collected and analyzed from adult C57Bl/6 male mice at four days and 12 weeks after folic acid-induced AKI. Mice were euthanized 12 weeks after AKI and structural metrics were obtained from cationic ferritin-enhanced-MRI (CFE-MRI) and histologic assessment. The fraction of proximal tubules, number of atubular glomeruli (ATG), and area of scarring were measured histologically. The correlation between the urinary biomarkers at the AKI or CKD and CFE-MRI derived features was determined, alone or in combination with the histologic features, using principal components. RESULTS: Using principal components derived from structural features, twelve urinary proteins were identified at the time of AKI that predicted structural changes 12 weeks after injury. The raw and normalized urinary concentrations of IGFBP-3 and TNFRII strongly correlated to the structural findings from histology and CFE-MRI. Urinary fractalkine concentration at the time of CKD correlated with structural findings of CKD. CONCLUSIONS: We have used structural features to identify several candidate urinary proteins that predict whole kidney pathologic features during the transition from AKI to CKD, including IGFBP-3, TNFRII, and fractalkine. In future work, these biomarkers must be corroborated in patient cohorts to determine their suitability to predict CKD after AKI.

Self-Supervised Contrastive Learning to Predict Alzheimer's Disease Progression with 3D Amyloid-PET.

Early diagnosis of Alzheimer's disease (AD) is an important task that facilitates the development of treatment and prevention strategies and may potentially improve patient outcomes. Neuroimaging has shown great promise, including the amyloid-PET which measures the accumulation of amyloid plaques in the brain - a hallmark of AD. It is desirable to train end-to-end deep learning models to predict the progression of AD for individuals at early stages based on 3D amyloid-PET. However, commonly used models are trained in a fully supervised learning manner and they are inevitably biased toward the given label information. To this end, we propose a self-supervised contrastive learning method to predict AD progression with 3D amyloid-PET. It uses unlabeled data to capture general representations underlying the images. As the downstream task is given as classification, unlike the general self-supervised learning problem that aims to generate task-agnostic representations, we also propose a loss function to utilize the label information in the pre-training. To demonstrate the performance of our method, we conducted experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. The results confirmed that the proposed method is capable of providing appropriate data representations, resulting in accurate classification.

Questionnaire and structural imaging data accurately predict headache improvement in patients with acute post-traumatic headache attributed to mild traumatic brain injury.

BACKGROUND: Our prior work demonstrated that questionnaires assessing psychosocial symptoms have utility for predicting improvement in patients with acute post-traumatic headache following mild traumatic brain injury. In this cohort study, we aimed to determine whether prediction accuracy can be refined by adding structural magnetic resonance imaging (MRI) brain measures to the model. METHODS: Adults with acute post-traumatic headache (enrolled 0-59 days post-mild traumatic brain injury) underwent T1-weighted brain MRI and completed three questionnaires (Sports Concussion Assessment Tool, Pain Catastrophizing Scale, and the Trait Anxiety Inventory Scale). Individuals with post-traumatic headache completed an electronic headache diary allowing for determination of headache improvement at three- and at six-month follow-up. Questionnaire and MRI measures were used to train prediction models of headache improvement and headache trajectory. RESULTS: Forty-three patients with post-traumatic headache (mean age = 43.0, SD = 12.4; 27 females/16 males) and 61 healthy controls were enrolled (mean age = 39.1, SD = 12.8; 39 females/22 males). The best model achieved cross-validation Area Under the Curve of 0.801 and 0.805 for predicting headache improvement at three and at six months. The top contributing MRI features for the prediction included curvature and thickness of superior, middle, and inferior temporal, fusiform, inferior parietal, and lateral occipital regions. Patients with post-traumatic headache who did not improve by three months had less thickness and higher curvature measures and notably greater baseline differences in brain structure vs. healthy controls (thickness: p < 0.001, curvature: p = 0.012) than those who had headache improvement. CONCLUSIONS: A model including clinical questionnaire data and measures of brain structure accurately predicted headache improvement in patients with post-traumatic headache and achieved improvement compared to a model developed using questionnaire data alone.

Multisite thalamic recordings to characterize seizure propagation in the human brain.

Neuromodulation of the anterior nuclei of the thalamus (ANT) has shown to be efficacious in a subset of patients with refractory focal epilepsy. One important uncertainty is to what extent thalamic subregions other than the ANT could be recruited more prominently in the propagation of focal onset seizures. We designed the current study to simultaneously monitor the engagement of the ANT, mediodorsal (MD) and pulvinar (PUL) nuclei during seizures in patients who could be candidates for thalamic neuromodulation. We studied 11 patients with clinical manifestations of presumed temporal lobe epilepsy (TLE) undergoing invasive stereo-encephalography (sEEG) monitoring to confirm the source of their seizures. We extended cortical electrodes to reach the ANT, MD and PUL nuclei of the thalamus. More than one thalamic subdivision was simultaneously interrogated in nine patients. We recorded seizures with implanted electrodes across various regions of the brain and documented seizure onset zones (SOZ) in each recorded seizure. We visually identified the first thalamic subregion to be involved in seizure propagation. Additionally, in eight patients, we applied repeated single pulse electrical stimulation in each SOZ and recorded the time and prominence of evoked responses across the implanted thalamic regions. Our approach for multisite thalamic sampling was safe and caused no adverse events. Intracranial EEG recordings confirmed SOZ in medial temporal lobe, insula, orbitofrontal and temporal neocortical sites, highlighting the importance of invasive monitoring for accurate localization of SOZs. In all patients, seizures with the same propagation network and originating from the same SOZ involved the same thalamic subregion, with a stereotyped thalamic EEG signature. Qualitative visual reviews of ictal EEGs were largely consistent with the quantitative analysis of the corticothalamic evoked potentials, and both documented that thalamic nuclei other than ANT could have the earliest participation in seizure propagation. Specifically, pulvinar nuclei were involved earlier and more prominently than ANT in more than half of the patients. However, which specific thalamic subregion first demonstrated ictal activity could not be reliably predicted based on clinical semiology or lobar localization of SOZs. Our findings document the feasibility and safety of bilateral multisite sampling from the human thalamus. This may allow more personalized thalamic targets to be identified for neuromodulation. Future studies are needed to determine if a personalized thalamic neuromodulation leads to greater improvements in clinical outcome.

Doppler radar remote sensing of respiratory function.

Doppler radar remote sensing of torso kinematics can provide an indirect measure of cardiopulmonary function. Motion at the human body surface due to heart and lung activity has been successfully used to characterize such measures as respiratory rate and depth, obstructive sleep apnea, and even the identity of an individual subject. For a sedentary subject, Doppler radar can track the periodic motion of the portion of the body moving as a result of the respiratory cycle as distinct from other extraneous motions that may occur, to provide a spatial temporal displacement pattern that can be combined with a mathematical model to indirectly assess quantities such as tidal volume, and paradoxical breathing. Furthermore, it has been demonstrated that even healthy respiratory function results in distinct motion patterns between individuals that vary as a function of relative time and depth measures over the body surface during the inhalation/exhalation cycle. Potentially, the biomechanics that results in different measurements between individuals can be further exploited to recognize pathology related to lung ventilation heterogeneity and other respiratory diagnostics.

Neuropsychiatric Symptoms and Commonly Used Biomarkers of Alzheimer's Disease: A Literature Review from a Machine Learning Perspective.

There is a growing interest in the application of machine learning (ML) in Alzheimer's disease (AD) research. However, neuropsychiatric symptoms (NPS), frequent in subjects with AD, mild cognitive impairment (MCI), and other related dementias have not been analyzed sufficiently using ML methods. To portray the landscape and potential of ML research in AD and NPS studies, we present a comprehensive literature review of existing ML approaches and commonly studied AD biomarkers. We conducted PubMed searches with keywords related to NPS, AD biomarkers, machine learning, and cognition. We included a total of 38 articles in this review after excluding some irrelevant studies from the search results and including 6 articles based on a snowball search from the bibliography of the relevant studies. We found a limited number of studies focused on NPS with or without AD biomarkers. In contrast, multiple statistical machine learning and deep learning methods have been used to build predictive diagnostic models using commonly known AD biomarkers. These mainly included multiple imaging biomarkers, cognitive scores, and various omics biomarkers. Deep learning approaches that combine these biomarkers or multi-modality datasets typically outperform single-modality datasets. We conclude ML may be leveraged to untangle the complex relationships of NPS and AD biomarkers with cognition. This may potentially help to predict the progression of MCI or dementia and develop more targeted early intervention approaches based on NPS.