Hasil untuk "Neurology. Diseases of the nervous system"

Ricardo Montoya Monsalve, Sara Isabel Castaño Ocampo

Introducción: la inteligencia artificial (IA) ha emergido como una herramienta relevante en neuropsicología, con potencial para optimizar procesos clínicos, investigativos y educativos en el estudio de las funciones cognitivas. Materiales y métodos: se realizó una revisión teórica mediante búsqueda en bases de datos como PubMed, Scopus, ScienceDirect, Scielo y Redalyc, empleando términos MeSH y operadores booleanos. Se incluyeron artículos en inglés y español, principalmente desde 2014. De 98 registros, se seleccionaron 52 tras aplicar criterios de inclusión y exclusión. Resultados: se identificaron tres áreas de aplicación: clínica, investigativa y educativa. En el ámbito clínico, la IA alcanzó precisiones de hasta el 91?% en la predicción de demencia y apoyó el análisis de neuroimágenes y el tratamiento. En investigación, facilitó el análisis de grandes volúmenes de datos, la identificación de biomarcadores y el desarrollo de modelos predictivos. en educación, mostró beneficios en el aprendizaje personalizado, aunque con menor nivel de evidencia. Discusión: persisten limitaciones como la baja interpretabilidad, problemas de generalización y desafíos éticos relacionados con sesgos y privacidad de datos. Conclusiones: la IA representa un avance significativo en neuropsicología, pero requiere marcos ético-legales y debe complementar, no sustituir, la experiencia clínica.

Muhammad Hammad Maqsood, Mubashir Sajid, Khubaib Ahmed et al.

This research paper outlines the development and implementation of a novel Clinical Decision Support System (CDSS) that integrates AI predictive modeling with medical knowledge bases. It utilizes the quantifiable information elements in lab results for inferring likely diagnoses a patient might have. Subsequently, suggesting investigations to confirm the likely diagnoses -- an assistive tool for physicians. The system fuses knowledge contained in a rule-base expert system with inferences of data driven predictors based on the features in labs. The data for 593,055 patients was collected from 547 primary care centers across the US to model our decision support system and derive Real-Word Evidence (RWE) to make it relevant for a large demographic of patients. Our Rule-Base comprises clinically validated rules, modeling 59 health conditions that can directly confirm one or more of diseases and assign ICD-10 codes to them. The Likely Diagnosis system uses multi-class classification, covering 37 ICD-10 codes, which are grouped together into 11 categories based on the labs that physicians prescribe to confirm the diagnosis. This research offers a novel system that assists a physician by utilizing medical profile of a patient and routine lab investigations to predict a group of likely diseases and then confirm them, coupled with providing explanations for inferences, thereby assisting physicians to reduce misdiagnosis of patients in clinical decision-making.

Volodymyr Sydorskyi, Igor Krashenyi, Oleksii Yakubenko

Melanoma detection is vital for early diagnosis and effective treatment. While deep learning models on dermoscopic images have shown promise, they require specialized equipment, limiting their use in broader clinical settings. This study introduces a multi-modal melanoma detection system using conventional photo images, making it more accessible and versatile. Our system integrates image data with tabular metadata, such as patient demographics and lesion characteristics, to improve detection accuracy. It employs a multi-modal neural network combining image and metadata processing and supports a two-step model for cases with or without metadata. A three-stage pipeline further refines predictions by boosting algorithms and enhancing performance. To address the challenges of a highly imbalanced dataset, specific techniques were implemented to ensure robust training. An ablation study evaluated recent vision architectures, boosting algorithms, and loss functions, achieving a peak Partial ROC AUC of 0.18068 (0.2 maximum) and top-15 retrieval sensitivity of 0.78371. Results demonstrate that integrating photo images with metadata in a structured, multi-stage pipeline yields significant performance improvements. This system advances melanoma detection by providing a scalable, equipment-independent solution suitable for diverse healthcare environments, bridging the gap between specialized and general clinical practices.

T. Roxana Ghadimi, Clare McCormack

The perinatal period represents a time of profound neurobiological, cognitive, and emotional change. While evidence points to the neuroplasticity of matrescence as adaptive in supporting the transition to motherhood, the perinatal period also entails subjective reports of cognitive difficulty known as “mommy brain” as well as a heightened vulnerability to mental health challenges. The role of cognition in the etiology of postpartum depression is a promising area of investigation into targets for maternal mental health intervention, considering evidence that important cognitive changes occur during the perinatal period, and given that cognitive alterations are key features of mood disorders. Here we review evidence for cognitive plasticity in matrescence, with a particular focus on executive function (EF) given its overlapping significance for adaptation to parenthood, central role in managing the mental load of motherhood, and implications in mood regulation and mood disorders. We also review evidence for EF changes in perinatal depression and major depressive disorder more broadly. Despite the strong association between EF impairments and major depressive disorder, research on EF changes in perinatal depression remains limited. Understanding normative EF changes during this period is essential for better understanding the relationship between EF, perinatal depression, and the mental load of motherhood. Consideration for these cognitive, neurobiological, and psychosocial factors of matrescence is critical for addressing maternal mental health and developing interventions that support parental well-being.

Sri Varsha Mulakala, G. Neeharika, P. Vinay Kumar et al.

The recent increase in morbidity is primarily due to chronic diseases including Diabetes, Heart disease, Lung cancer, and brain tumours. The results for patients can be improved, and the financial burden on the healthcare system can be lessened, through the early detection and prevention of certain disorders. In this study, we built a machine-learning model for predicting the existence of numerous diseases utilising datasets from various sources, including Kaggle, Dataworld, and the UCI repository, that are relevant to each of the diseases we intended to predict. Following the acquisition of the datasets, we used feature engineering to extract pertinent features from the information, after which the model was trained on a training set and improved using a validation set. A test set was then used to assess the correctness of the final model. We provide an easy-to-use interface where users may enter the parameters for the selected ailment. Once the right model has been run, it will indicate whether the user has a certain ailment and offer suggestions for how to treat or prevent it.

Ran Piao, Yuan Lu, Hareld Kemps et al.

Voice-based health assessment offers unprecedented opportunities for scalable, non-invasive disease screening, yet existing approaches typically focus on single conditions and fail to leverage the rich, multi-faceted information embedded in speech. We present MARVEL (Multi-task Acoustic Representations for Voice-based Health Analysis), a privacy-conscious multitask learning framework that simultaneously detects nine distinct neurological, respiratory, and voice disorders using only derived acoustic features, eliminating the need for raw audio transmission. Our dual-branch architecture employs specialized encoders with task-specific heads sharing a common acoustic backbone, enabling effective cross-condition knowledge transfer. Evaluated on the large-scale Bridge2AI-Voice v2.0 dataset, MARVEL achieves an overall AUROC of 0.78, with exceptional performance on neurological disorders (AUROC = 0.89), particularly for Alzheimer's disease/mild cognitive impairment (AUROC = 0.97). Our framework consistently outperforms single-modal baselines by 5-19% and surpasses state-of-the-art self-supervised models on 7 of 9 tasks, while correlation analysis reveals that the learned representations exhibit meaningful similarities with established acoustic features, indicating that the model's internal representations are consistent with clinically recognized acoustic patterns. By demonstrating that a single unified model can effectively screen for diverse conditions, this work establishes a foundation for deployable voice-based diagnostics in resource-constrained and remote healthcare settings.

Muhammad Ali Bangash, Julie Qiaojin Lin

Aims Local protein synthesis at the synapse is a key determinant of learning and memory and is predicted to be severely disrupted in Alzheimer's disease (AD). Omics approaches have played a key role in deciphering molecular mechanisms underlying AD pathology. However, isolating the transcriptome may be biased due to inherent variations in transcript levels, or by transcription-on-demand models employed by several genes, whereas mass-spec based proteomics approaches fail to capture low abundance peptides. The translatome bypasses these inherent limitations of other omics methods by capturing actively translating mRNA species trapped inside ribosomes and subjecting them to unbiased RNA-seq analysis capturing even very low abundance transcripts. Methods Isolating the neuronal ribosomes from human post-mortem brains without interference from non-neuronal cells remains a challenge. We used frozen brain tissue from Alzheimer's patients and healthy controls obtained from the Cambridge Brain Biobank. Synaptoneurosomal fractions were prepared using sucrose gradients in non-denaturing buffers with RNAse inhibitors to preserve ribosomal composition and trapped mRNA. We isolated functional ribosomes on affinity columns following recombinant RNAse digestion. Finally, actively translating ribosome-trapped mRNAs were sequenced using RNA-seq, aligned to human genome using STAR alignment and analysed for differential expression using DeSeq2 followed by pathway analysis. Results We have successfully isolated ribosome-associated RNA transcripts in the dendritic spines from cortical neurons of postmortem Alzheimer's brains with little interference from glial and non-neuronal material. The novel AD translatome disruptions identified by isolating endogenous ribosome bound mRNA will help detect downstream molecular targets. We will also integrate targeted translatome data with published transcriptome and GWAS DNA variant data to identify novel biomarkers. Conclusion This is the first successful isolation of the dendritic translatome from human postmortem AD brains. Future studies will verify functional significance of key targets using gain- and loss-of-function studies in animal models of AD and human iPSCs.

Zhang Y, Wang H, Yang J et al.

Yating Zhang,1 Hongyan Wang,1 Jie Yang,2 Sanchun Wang,1 Weifang Tong,1 Bo Teng1 1Department of Otorhinolaryngology Head and Neck Surgery, the Second Hospital of Jilin University, Changchun, Jilin Province, People’s Republic of China; 2Department of Neurology, the First Hospital of Jilin University, Changchun, Jilin Province, People’s Republic of ChinaCorrespondence: Bo Teng, Department of Otorhinolaryngology Head and Neck Surgery, the Second Hospital of Jilin University, No. 218 Ziqiang Street, Nanguan District, Changchun, Jilin Province, 130000, People’s Republic of China, Email tengbo1975@163.comPurpose: This investigation sought to elucidate the genetic underpinnings that connect obesity indicators, circulating blood lipid levels, adipokines levels and obstructive sleep apnea syndrome (OSAS), employing a bidirectional two-sample Mendelian randomization (MR) analysis that utilizes data derived from extensive genome-wide association studies (GWAS).Methods: We harnessed genetic datasets of OSAS available from the FinnGen consortium and summary data of four obesity indices (including neck circumference), seven blood lipid (including triglycerides) and eleven adipokines (including leptin) from the IEU OpenGWAS database. We primarily utilized inverse variance weighted (IVW), weighted median, and MR-Egger methods, alongside MR-PRESSO and Cochran’s Q tests, to validate and assess the diversity and heterogeneity of our findings.Results: After applying the Bonferroni correction, we identified significant correlations between OSAS and increased neck circumference (Odds Ratio [OR]: 3.472, 95% Confidence Interval [CI]: 1.954– 6.169, P= 2.201E-05) and decreased high-density lipoprotein (HDL) cholesterol levels (OR: 0.904, 95% CI: 0.858– 0.952, P= 1.251E-04). Concurrently, OSAS was linked to lower leptin levels (OR: 1.355, 95% CI: 1.069– 1.718, P= 0.012) and leptin receptor levels (OR: 0.722, 95% CI: 0.530– 0.996, P= 0.047). Sensitivity analyses revealed heterogeneity in HDL cholesterol and leptin indicators, but further multiplicative random effects IVW method analysis confirmed these correlations as significant (P< 0.05) without notable heterogeneity or horizontal pleiotropy in other instrumental variables.Conclusion: This investigation compellingly supports the hypothesis that OSAS could be a genetic predisposition for elevated neck circumference, dyslipidemia, and adipokine imbalance. These findings unveil potential genetic interactions between OSAS and metabolic syndrome, providing new pathways for research in this domain. Future investigations should aim to delineate the specific biological pathways by which OSAS impacts metabolic syndrome. Understanding these mechanisms is critical for developing targeted prevention and therapeutic strategies.Keywords: sleep disorders, metabolic syndrome, causal inference, GWAS

Ahmed Akib Jawad Karim, Muhammad Zawad Mahmud, Samiha Islam et al.

In this research, we explored the improvement in terms of multi-class disease classification via pre-trained language models over Medical-Abstracts-TC-Corpus that spans five medical conditions. We excluded non-cancer conditions and examined four specific diseases. We assessed four LLMs, BioBERT, XLNet, and BERT, as well as a novel base model (Last-BERT). BioBERT, which was pre-trained on medical data, demonstrated superior performance in medical text classification (97% accuracy). Surprisingly, XLNet followed closely (96% accuracy), demonstrating its generalizability across domains even though it was not pre-trained on medical data. LastBERT, a custom model based on the lighter version of BERT, also proved competitive with 87.10% accuracy (just under BERT's 89.33%). Our findings confirm the importance of specialized models such as BioBERT and also support impressions around more general solutions like XLNet and well-tuned transformer architectures with fewer parameters (in this case, LastBERT) in medical domain tasks.

Jixun Yao, Nikita Kuzmin, Qing Wang et al.

Speaker anonymization is an effective privacy protection solution that conceals the speaker's identity while preserving the linguistic content and paralinguistic information of the original speech. To establish a fair benchmark and facilitate comparison of speaker anonymization systems, the VoicePrivacy Challenge (VPC) was held in 2020 and 2022, with a new edition planned for 2024. In this paper, we describe our proposed speaker anonymization system for VPC 2024. Our system employs a disentangled neural codec architecture and a serial disentanglement strategy to gradually disentangle the global speaker identity and time-variant linguistic content and paralinguistic information. We introduce multiple distillation methods to disentangle linguistic content, speaker identity, and emotion. These methods include semantic distillation, supervised speaker distillation, and frame-level emotion distillation. Based on these distillations, we anonymize the original speaker identity using a weighted sum of a set of candidate speaker identities and a randomly generated speaker identity. Our system achieves the best trade-off of privacy protection and emotion preservation in VPC 2024.

Aymeé Hernández Hernández

La polineuropatia desmielinizante inflamatoria crónica (CIDP), por sus siglas en inglés, es una polineurorradicu-lopatía predominantemente motora de etiología autoinmune. Su comienzo es insidioso y el curso crónico, afecta tanto a hombres como a mujeres con un ligero predominio en los primeros. Puede aparecer en cualquier etapa de la vida, es más frecuente a partir de la segunda década. Es necesario diferenciarla de otras enfermedades, ya que es potencialmente tratable, con buena respuesta a los inmunosupresores. Su diagnóstico se sustenta en cuatro pilares: la clínica, los estudios neurofisiológicos, el estudio del líquido cefalorraquideo y los análisis anatomopatológicos. Desde su descripción inicial se han planteado numerosos criterios diagnósticos, aspecto polémico y controversial, ya que en la práctica clínica existen numerosos pacientes que no cumplen estrictamente con estos criterios pues presentan casos atípicos. Por la dificultad en el diagnóstico de esta patología se realizó una revisión del tema, exponiendo los principales aspectos clínicos, electrofisiológicos, fisiopatológicos y anatomopatológicos, junto con una breve exposición de lo hallado en doce años de experiencia con esta patología en la Habana, Cuba.

E. Owusu, N. Nkire, F. Oluwasina et al.

Introduction Appropriate and adequate treatment of psychiatric conditions in the community or at first presentation to the hospital may prevent rehospitalization. Information about hospital readmission factors may help to reduce readmission rates. Objectives The scoping review sought to examine the readmission of patients to acute psychiatric hospitals to determine predictors and interventions to reduce psychiatric readmission rates. Methods A scoping review was conducted in eleven bibliographic databases to identify the relevant peer‐reviewed studies. Two reviewers independently assessed full‐text articles, and a screening process was undertaken to identify studies for inclusion in the review. PRISMA checklist was adopted, and with the Covidence software, 75 articles were eligible for review. Data extraction was conducted, collated, summarized, and findings were reported. Results The outcome of the review shows that learning disabilities, developmental delays, and alcohol, drug, and substance abuse, were crucial factors that increased the risk of readmission. It was also established through the review that greater access to mental health services in residential treatment and improved crisis intervention in congregate care settings were indicated as factors that reduce the risk of readmission. Conclusions High rates of readmission may adversely impact healthcare spending. This study suggests a need for focused health policies to address readmission factors and improve community‐based care. Disclosure of Interest None Declared

Steve Chung, Jurriaan M. Peters, Kamil Detyniecki et al.

Nasal administration of treatments for neurologic conditions, including rescue therapies to treat seizure clusters among people with epilepsy, represents a meaningful advance in patient care. Nasal anatomy and physiology underpin the multiple advantages of nasal administration but also present challenges that must be addressed in any successful nasal formulation. Nasal cavity anatomy is complex, with a modest surface area for absorption that limits the dose volume of an intranasal formulation. The mucociliary clearance mechanism and natural barriers of the nasal epithelia must be overcome for adequate absorption. An extensive vasculature and the presence of olfactory nerves in the nasal cavity enable both systemic and direct-to-brain delivery of drugs targeting the central nervous system. Two intranasal benzodiazepine rescue therapies have been approved by the US Food and Drug Administration for seizure-cluster treatment, in addition to the traditional rectal formulation. Nasal sprays are easy to use and offer the potential for quick and consistent bioavailability. This review aims to increase the clinician’s understanding of nasal anatomy and physiology and of the formulation of intranasal rescue therapies and to facilitate patient education and incorporate intranasal rescue therapies for seizure clusters (also known as acute repetitive seizures) into their seizure action plans.

Natália de Oliveira Ferrarini , Izabely Lima Assunção, Márcia Andréa Silva Carvalho Sombra et al.

Evidências crescentes sugerem que a farmacoterapia pode ser benéfica para alguns pacientes com transtorno da compulsão alimentar, um transtorno alimentar caracterizado por episódios repetitivos de consumo incontrolável de quantidades anormalmente grandes de alimentos sem comportamentos inadequados de perda de peso. Diante disso, este estudo teve como objetivo avaliar a eficácia de alternativas terapêuticas farmacológicas no tratamento do transtorno da compulsão alimentar. Assim, realizou-se uma revisão sistemática a partir da seleção de estudos científicos publicados nos anos de 2017 a 2022. Com base na análise e interpretação dos dados, concluiu-se que alternativas terapêuticas farmacológicas são recursos complementares tanto no tratamento do transtorno da compulsão alimentar como de sintomas de desordem alimentar e não substitutas. Nesse sentido, o uso de medicamentos tais como fluoxetina, lisdexamfetamina e simplicifolia, aliado a outros tratamentos, como a psicoterapia, podem ser eficazes para pacientes e suas necessidades específicas.

C. Bayam, M. Tomé, C. Pedro et al.

Introduction Autism is a neurodevelopmental disorder characterized by deficits in the ability to initiate and maintain social interaction, as well as a set of restricted and inflexible behavior patterns and interests. Individuals with Autism Spectrum Disorder (ASD) are at increased risk of suicidal behavior, including suicidal ideation, suicide attempts and death by suicide, as compared to the general population. Among the underlying causes, the co-occurrence of other psychiatric disorders, such as depression and anxiety, is common and can contribute to the reduction of the quality of life, as well as a worse prognosis of the disease. Objectives Case report and brief review of risk factors associated with suicidal behavior in individuals with ASD. Methods Review of the patients clinical file; Brief non-sistematic literature review of articles indexed to Pubmed with the key words: “Autism Spectrum Disorder”, “Suicide”, ”Suicidal behaviour”, ”Mood disorder”. Results J., 18 years old, male, with ASD, the best student at school, with above-average results since childhood. Two years ago he showed a non-reciprocal love interest. Since then, he has had multiple visits to the emergency department and successive hospitalizations, mostly because of mood and behaviour alterations, with suicidal ideation. After 1 month with depressive and anxious symptoms, he ended up making a suicide attempt through voluntary intoxication by prescribed medication. He was taken to the emergency room. Examination of mental status highlighted depressed mood, elevated anxiety levels, hypoprosody, and active suicidal ideation. Blood tests and CE-CT scan without changes. He was admitted in the psychiatry ward and treated with fluvoxamine, risperidone and lorazepam. He showed a good evolution of the psychopathological condition. Discharged at day 44, he was referred to a psychiatric and psychological outpatient clinics. Conclusions Mood disorders have a significant impact on the well-being of individuals with ASD, contributing to a worse quality of life and higher suicide mortality. Cognition has been associated with different levels of death by suicide, and individuals with ASD without intellectual disability, such as this patient, are at increased risk of suicide, which may be due to a greater awareness of their own difficulties. The role of genetics has been a subject of interest. The overlap of genes strongly associated with suicidal behavior and ASD has been described. However, there is still need of large scale genetic studies, for a better understanding of the genetic mechanisms involved in this association. The identification of vulnerable individuals and early initiation of preventive and therapeutic strategies is essential to improve the prognosis of ASD. Disclosure of Interest None Declared

Wentao Zhang, Yujun Huang, Tong Zhang et al.

Currently intelligent diagnosis systems lack the ability of continually learning to diagnose new diseases once deployed, under the condition of preserving old disease knowledge. In particular, updating an intelligent diagnosis system with training data of new diseases would cause catastrophic forgetting of old disease knowledge. To address the catastrophic forgetting issue, an Adapter-based Continual Learning framework called ACL is proposed to help effectively learn a set of new diseases at each round (or task) of continual learning, without changing the shared feature extractor. The learnable lightweight task-specific adapter(s) can be flexibly designed (e.g., two convolutional layers) and then added to the pretrained and fixed feature extractor. Together with a specially designed task-specific head which absorbs all previously learned old diseases as a single "out-of-distribution" category, task-specific adapter(s) can help the pretrained feature extractor more effectively extract discriminative features between diseases. In addition, a simple yet effective fine-tuning is applied to collaboratively fine-tune multiple task-specific heads such that outputs from different heads are comparable and consequently the appropriate classifier head can be more accurately selected during model inference. Extensive empirical evaluations on three image datasets demonstrate the superior performance of ACL in continual learning of new diseases. The source code is available at https://github.com/GiantJun/CL_Pytorch.

Haider Sultan, Hafiza Farwa Mahmood, Noor Fatima et al.

Like other fields of Traditional Medicines, Unani Medicines have been found as an effective medical practice for ages. It is still widely used in the subcontinent, particularly in Pakistan and India. However, Unani Medicines Practitioners are lacking modern IT applications in their everyday clinical practices. An Online Clinical Decision Support System may address this challenge to assist apprentice Unani Medicines practitioners in their diagnostic processes. The proposed system provides a web-based interface to enter the patient's symptoms, which are then automatically analyzed by our system to generate a list of probable diseases. The system allows practitioners to choose the most likely disease and inform patients about the associated treatment options remotely. The system consists of three modules: an Online Clinical Decision Support System, an Artificial Intelligence Inference Engine, and a comprehensive Unani Medicines Database. The system employs advanced AI techniques such as Decision Trees, Deep Learning, and Natural Language Processing. For system development, the project team used a technology stack that includes React, FastAPI, and MySQL. Data and functionality of the application is exposed using APIs for integration and extension with similar domain applications. The novelty of the project is that it addresses the challenge of diagnosing diseases accurately and efficiently in the context of Unani Medicines principles. By leveraging the power of technology, the proposed Clinical Decision Support System has the potential to ease access to healthcare services and information, reduce cost, boost practitioner and patient satisfaction, improve speed and accuracy of the diagnostic process, and provide effective treatments remotely. The application will be useful for Unani Medicines Practitioners, Patients, Government Drug Regulators, Software Developers, and Medical Researchers.

T. V. Afanasieva, A. P. Kuzlyakin, A. V. Komolov

Cardiovascular diseases are widespread among patients with chronic noncommunicable diseases and are one of the leading causes of death, including in the working age. The article presents the relevance of the development and application of patient-oriented systems, in which machine learning (ML) is a promising technology that allows predicting cardiovascular diseases. Automated machine learning (AutoML) makes it possible to simplify and speed up the process of developing AI/ML applications, which is key in the development of patient-oriented systems by application users, in particular medical specialists. The authors propose a framework for the application of automatic machine learning and three scenarios that allowed for data combining five data sets of cardiovascular disease indicators from the UCI Machine Learning Repository to investigate the effectiveness in detecting this class of diseases. The study investigated one AutoML model that used and optimized the hyperparameters of thirteen basic ML models (KNeighborsUnif, KNeighborsDist, LightGBMXT, LightGBM, RandomForestGini, RandomForestEntr, CatBoost, ExtraTreesGini, ExtraTreesEntr, NeuralNetFastA, XGBoost, NeuralNetTorch, LightGBMLarge) and included the most accurate models in the weighted ensemble. The results of the study showed that the structure of the AutoML model for detecting cardiovascular diseases depends not only on the efficiency and accuracy of the basic models used, but also on the scenarios for preprocessing the initial data, in particular, on the technique of data normalization. The comparative analysis showed that the accuracy of the AutoML model in detecting cardiovascular disease varied in the range from 87.41% to 92.3%, and the maximum accuracy was obtained when normalizing the source data into binary values, and the minimum was obtained when using the built-in AutoML technique.

Eunsong Kang, Da-woon Heo, Jiwon Lee et al.

To understand the biological characteristics of neurological disorders with functional connectivity (FC), recent studies have widely utilized deep learning-based models to identify the disease and conducted post-hoc analyses via explainable models to discover disease-related biomarkers. Most existing frameworks consist of three stages, namely, feature selection, feature extraction for classification, and analysis, where each stage is implemented separately. However, if the results at each stage lack reliability, it can cause misdiagnosis and incorrect analysis in afterward stages. In this study, we propose a novel unified framework that systemically integrates diagnoses (i.e., feature selection and feature extraction) and explanations. Notably, we devised an adaptive attention network as a feature selection approach to identify individual-specific disease-related connections. We also propose a functional network relational encoder that summarizes the global topological properties of FC by learning the inter-network relations without pre-defined edges between functional networks. Last but not least, our framework provides a novel explanatory power for neuroscientific interpretation, also termed counter-condition analysis. We simulated the FC that reverses the diagnostic information (i.e., counter-condition FC): converting a normal brain to be abnormal and vice versa. We validated the effectiveness of our framework by using two large resting-state functional magnetic resonance imaging (fMRI) datasets, Autism Brain Imaging Data Exchange (ABIDE) and REST-meta-MDD, and demonstrated that our framework outperforms other competing methods for disease identification. Furthermore, we analyzed the disease-related neurological patterns based on counter-condition analysis.

Fernando L. Pagan, Yasar Torres‐Yaghi, Michaeline L. Hebron et al.

Abstract Introduction Bosutinib, a dual Abelson/Src inhibitor, was investigated in individuals with dementia with Lewy bodies (DLB). Methods A single site, randomized, double‐blind, placebo‐controlled study of the effects of oral bosutinib, 100 mg once daily for 12 weeks on primary safety and pharmacokinetics and secondary biomarker outcomes. Results Twenty‐six participants were randomized and included male and female (12:1) in the bosutinib arm and all male (13) in the placebo arm. The average age was 72.9 ± 8.1 (year ± standard deviation). There were no serious adverse events and no dropouts. Bosutinib was measured in the cerebrospinal fluid (CSF) and inhibited Abelson. Bosutinib reduced CSF alpha‐synuclein and dopamine catabolism. Discussion Bosutinib is safe and well tolerated and penetrates the blood–brain barrier to inhibit Abelson and reduce CSF alpha‐synuclein and dopamine catabolism, suggesting that bosutinib (100 mg) may be at or near the lowest effective dose in DLB. These results will guide adequately powered studies to determine the efficacy of a dose range of bosutinib and longer treatment in DLB. Highlights Bosutinib is a dual Abl/Src inhibitor that penetrates the blood brain barrier Bosutinib is safe and tolerated in individuals with dementia with Lewy bodies Bosutinib engages its target via inhibition of Abl and Src Bosutinib reduces CSF alpha‐synuclein and attenuates breakdown of dopamine Bosutinib improves activities of daily living in dementia with Lewy bodies