Identifying gray matter alterations in Cushing's disease using machine learning: An interpretable approach.

BACKGROUND: Cushing's Disease (CD) is a rare clinical syndrome characterized by excessive secretion of adrenocorticotrophic hormone, leading to significant functional and structural brain alterations as observed in Magnetic Resonance Imaging (MRI). While traditional statistical analysis has been widely employed to investigate these MRI changes in CD, it has lacked the ability to predict individual-level outcomes. PURPOSE: To address this problem, this paper has proposed an interpretable machine learning (ML) framework, including model-level assessment, feature-level assessment, and biology-level assessment to ensure a comprehensive analysis based on structural MRI of CD. METHODS: The ML framework has effectively identified the changes in brain regions in the stage of model-level assessment, verified the effectiveness of these altered brain regions to predict CD from normal controls in the stage of feature-level assessment, and carried out a correlation analysis between altered brain regions and clinical symptoms in the stage of biology-level assessment. RESULTS: The experimental results of this study have demonstrated that the Insula, Fusiform gyrus, Superior frontal gyrus, Precuneus, and the opercular portion of the Inferior frontal gyrus of CD showed significant alterations in brain regions. Furthermore, our study has revealed significant correlations between clinical symptoms and the frontotemporal lobes, insulin, and olfactory cortex, which also have been confirmed by previous studies. CONCLUSIONS: The ML framework proposed in this study exhibits exceptional potential in uncovering the intricate pathophysiological mechanisms underlying CD, with potential applicability in diagnosing other diseases.

Brain Structural Connectivity Guided Vision Transformers for Identification of Functional Connectivity Characteristics in Preterm Neonates.

Preterm birth is the leading cause of death in children under five years old, and is associated with a wide sequence of complications in both short and long term. In view of rapid neurodevelopment during the neonatal period, preterm neonates may exhibit considerable functional alterations compared to term ones. However, the identified functional alterations in previous studies merely achieve moderate classification performance, while more accurate functional characteristics with satisfying discrimination ability for better diagnosis and therapeutic treatment is underexplored. To address this problem, we propose a novel brain structural connectivity (SC) guided Vision Transformer (SCG-ViT) to identify functional connectivity (FC) differences among three neonatal groups: preterm, preterm with early postnatal experience, and term. Particularly, inspired by the neuroscience-derived information, a novel patch token of SC/FC matrix is defined, and the SC matrix is then adopted as an effective mask into the ViT model to screen out input FC patch embeddings with weaker SC, and to focus on stronger ones for better classification and identification of FC differences among the three groups. The experimental results on multi-modal MRI data of 437 neonatal brains from publicly released Developing Human Connectome Project (dHCP) demonstrate that SCG-ViT achieves superior classification ability compared to baseline models, and successfully identifies holistically different FC patterns among the three groups. Moreover, these different FCs are significantly correlated with the differential gene expressions of the three groups. In summary, SCG-ViT provides a powerfully brain-guided pipeline of adopting large-scale and data-intensive deep learning models for medical imaging-based diagnosis.

Case report: Excessive daytime sleepiness as a presenting manifestation of autoimmune glial fibrillary acidic protein astrocytopathy.

Autoimmune glial fibrillary acidic protein astrocytopathy (GFAP-A) is a recently discovered autoimmune inflammatory disease of the central nervous system. It presents with a variety of clinical symptoms, including fever, seizures, psychiatric symptoms, limber weakness, and sensory symptoms. However, the symptoms of sleep disorders have not been sufficiently addressed. Here, we report a case of GFAP-A in which the patient complained of excessive daytime sleepiness and an excessive need for sleep. Our patient was a 58-year-old male who experienced excessive daytime sleepiness for 50 days following SARS-CoV-2 infection. He was diagnosed with coronavirus disease 2019 on June 1st. On the 7th of June, he experienced excessive daytime sleepiness, nausea, reduced food intake, lower limb weakness, and dysuria. Subsequently, his sleepiness significantly deteriorated on July 21st. Five months prior, the patient underwent laparoscopic partial right nephrectomy for clear-cell renal cell carcinoma. Brain MRI revealed abnormal hyperintense lesions in the pontine brain and around the mesencephalic aqueduct on T2 and T2-fluid attenuated inversion recovery (T2-FLAIR) sequences However, these lesions did not exhibit any pathological enhancement. Spinal cord MRI revealed lesions in the C6-C7 and T2-T3 segments on the T2 sequence. His Epworth Sleepiness Scale (ESS) score was 16 (reference range, <10), and 24-hour polysomnography supported the diagnosis of rapid-eye-movement sleep disorder and severe sleep apnea-hypopnea syndrome. Glial fibrillary acidic protein IgG antibodies were detected in the cerebrospinal fluid (1:32, cell-based assay) but not in the serum. The level of hypocretin in the cerebrospinal fluid was 29.92 pg/mL (reference range ≥110 pg/mL), suggesting narcolepsy type 1. After treatment with corticosteroids for approximately 1 month, the patient showed considerable clinical and radiological improvement, as well as an increase in hypocretin levels. Although repeated polysomnography and multiple sleep latency tests suggested narcolepsy, his ESS score decreased to 8. Our findings broaden the range of clinical manifestations associated with GFAP-A, thereby enhancing diagnostic and therapeutic strategies for this disease. Additionally, our results indicate a potential common autoimmune mechanism involving GFAP-A and orexin system dysregulation, warranting further investigation.

A brain structural connectivity biomarker for autism spectrum disorder diagnosis in early childhood.

Background: Autism spectrum disorder (ASD) is associated with altered brain development, but it is unclear which specific structural changes may serve as potential diagnostic markers, particularly in young children at the age when symptoms become fully established. Furthermore, such brain markers need to meet the requirements of precision medicine and be accurate in aiding diagnosis at an individual rather than only a group level. Objective: This study aimed to identify and model brain-wide differences in structural connectivity using diffusion tensor imaging (DTI) in young ASD and typically developing (TD) children. Methods: A discovery cohort including 93 ASD and 26 TD children and two independent validation cohorts including 12 ASD and 9 TD children from three different cities in China were included. Brain-wide (294 regions) structural connectivity was measured using DTI (fractional anisotropy, FA) together with symptom severity and cognitive development. A connection matrix was constructed for each child for comparisons between ASD and TD groups. Pattern classification was performed on the discovery dataset and the resulting model was tested on the two independent validation datasets. Results: Thirty-three structural connections showed increased FA in ASD compared to TD children and associated with both autistic symptom severity and impaired general cognitive development. The majority (29/33) involved the frontal lobe and comprised five different networks with functional relevance to default mode, motor control, social recognition, language and reward. Overall, classification achieved very high accuracy of 96.77% in the discovery dataset, and 91.67% and 88.89% in the two independent validation datasets. Conclusions: Identified structural connectivity differences primarily involving the frontal cortex can very accurately distinguish novel individual ASD from TD children and may therefore represent a robust early brain biomarker which can address the requirements of precision medicine.

Deep Neural Networks for Depression Recognition Based on 2D and 3D Facial Expressions Under Emotional Stimulus Tasks.

The proportion of individuals with depression has rapidly increased along with the growth of the global population. Depression has been the currently most prevalent mental health disorder. An effective depression recognition system is especially crucial for the early detection of potential depression risk. A depression-related dataset is also critical while evaluating the system for depression or potential depression risk detection. Due to the sensitive nature of clinical data, availability and scale of such datasets are scarce. To our knowledge, there are few extensively practical depression datasets for the Chinese population. In this study, we first create a large-scale dataset by asking subjects to perform five mood-elicitation tasks. After each task, subjects' audio and video are collected, including 3D information (depth information) of facial expressions via a Kinect. The constructed dataset is from a real environment, i.e., several psychiatric hospitals, and has a specific scale. Then we propose a novel approach for potential depression risk recognition based on two kinds of different deep belief network (DBN) models. One model extracts 2D appearance features from facial images collected by an optical camera, while the other model extracts 3D dynamic features from 3D facial points collected by a Kinect. The final decision result comes from the combination of the two models. Finally, we evaluate all proposed deep models on our built dataset. The experimental results demonstrate that (1) our proposed method is able to identify patients with potential depression risk; (2) the recognition performance of combined 2D and 3D features model outperforms using either 2D or 3D features model only; (3) the performance of depression recognition is higher in the positive and negative emotional stimulus, and females' recognition rate is generally higher than that for males. Meanwhile, we compare the performance with other methods on the same dataset. The experimental results show that our integrated 2D and 3D features DBN is more reasonable and universal than other methods, and the experimental paradigm designed for depression is reasonable and practical.

Dynamic Cerebral Autoregulation in Embolic Stroke of Undetermined Source.

BACKGROUND AND PURPOSE: Dynamic cerebral autoregulation (dCA) in acute ischemic stroke is probably compromised. Although the characteristics of dCA in different types of stroke have been largely investigated, dCA in embolic stroke of undetermined source (ESUS) remains poorly understood. In this group, we aimed to elucidate the characteristics of dCA and their relevance to clinical outcomes. METHODS: The study enrolled 77 ESUS patients and 50 controls. Bilateral cerebral blood flow velocities (CBFV) of middle cerebral arteries and arterial blood pressure were simultaneously recorded using a transcranial Doppler combined with a servo-controlled finger plethysmograph. Transfer function analysis was used to obtain dCA parameters including phase, gain, coherence at very low frequency (VLF) and low frequency (LF), and the rate of recovery (RoRc) of CBFV. A multivariable logistic regression model was established to explore the relationship between dCA and clinical outcomes. RESULTS: Gain at VLF and LF, phase at LF, and RoRc of CBFV in bilateral hemispheres of the ESUS group were consistently worse than those of the control group (all P < 0.001). Bilateral RoRc of CBFV was significantly higher in patients with favorable outcomes than in those with unfavorable outcomes (stroke hemisphere: P < 0.001; non-stroke hemisphere, P = 0.029). Rate of recovery of CBFV in stroke hemisphere >13.3%/s was an independent predictor of favorable clinical outcomes (adjusted odds ratio = 30.95, 95% CI: 5.33-179.81, P < 0.001). CONCLUSIONS: Dynamic cerebral autoregulation was relatively impaired in both stroke and non-stroke hemispheres in ESUS patients, and functioning dCA after ESUS may indicate favorable clinical outcomes.

Changes in cerebral autoregulation and blood biomarkers after remote ischemic preconditioning.

OBJECTIVE: To determine the effect of remote ischemic preconditioning (RIPC) on dynamic cerebral autoregulation (dCA) and various blood biomarkers in healthy adults. METHODS: A self-controlled interventional study was conducted. Serial measurements of dCA were performed at 7 time points (7, 9, and 11 am; 2, 5, and 8 pm, and 8 am on the next day) without or with RIPC, carried out at 7:20 to 8 am. Venous blood samples were collected at baseline (7 am) and 1 hour after RIPC, and blood biomarkers, including 5 neuroprotective factors and 25 inflammation-related biomarkers, were measured with a quantitative protein chip. RESULTS: Fifty participants were enrolled (age 34.54 ± 12.01 years, 22 men). Compared with the results on the day without RIPC, dCA was significantly increased at 6 hours after RIPC, and the increase was sustained for at least 24 hours. After RIPC, 2 neuroprotective factors (glial cell-derived neurotrophic factor and vascular endothelial growth factor-A) and 4 inflammation-related biomarkers (transforming growth factor-ß1, leukemia inhibitory factor, matrix metallopeptidase-9, and tissue inhibitor of metalloproteinase-1) were significantly elevated compared with their baseline levels. Conversely, monocyte chemoattractant protein-1 was significantly lower compared with its baseline level. CONCLUSIONS: RIPC induces a sustained increase of dCA from 6 to at least 24 hours after treatment in healthy adults. In addition, several neuroprotective and inflammation-related blood biomarkers were differentially regulated shortly after RIPC. The increased dCA and altered blood biomarkers may collectively contribute to the beneficial effects of RIPC on cerebrovascular function. CLINICALTRIALSGOV IDENTIFIER: NCT02965547.

Dynamic Cerebral Autoregulation Remains Stable During the Daytime (8 a.m. to 8 p.m.) in Healthy Adults.

Many functions of the human body possess a daily rhythm, disruptions of which often lead to disease. Dynamic cerebral autoregulation (dCA) stabilizes the cerebral blood flow to prompt normal neural function. However, whether dCA is stable across the day remains unknown. This study aimed to investigate the daily rhythm of dCA. Fifty-one healthy adults (38.294 ± 13.279 years, 40 females) were recruited and received six dCA measurements per individual that were conducted at predefined time points: 8:00, 9:00, 11:00, 14:00, 17:00, and 20:00. Although the blood pressure fluctuated significantly, there was no statistical difference in phase difference and gain (autoregulatory parameters) across the six time points. This study demonstrates that dCA remains stable during the interval from 8 a.m. to 8 p.m. and underscores the importance of stable dCA in maintaining cerebral blood flow and neural function.

Hematoma volume is a predictive factor of disturbed autoregulation after spontaneous intracerebral hemorrhage.

BACKGROUND: Dynamic cerebral autoregulation (dCA) is probably disturbed after spontaneous intracerebral hemorrhage (ICH) and could be an independent predictor of clinical outcome. Clinical determinants of dCA status after ICH need to be further elucidated. METHODS: We finally enrolled 53 patients diagnosed with supratentorial ICH within 6days from onset. DCA was assessed 4-6days after ICH onset by monitoring middle cerebral artery blood flow velocities and simultaneous arterial blood pressure continuously, utilizing transcranial Doppler combined with a servo-controlled finger plethysmograph. Cerebral autoregulation was evaluated by phase difference (PD) in low frequency (0.06-0.12Hz) range derived from transfer function analysis. The previous clinical history was collected including hypertension, diabetes mellitus, prior stroke, smoking, heavy drinking history. Laboratory results during hospitalization were utilized for further risk factors screening, including fasting blood glucose, glycosylated hemoglobin A (1C), total cholesterol, low density lipoprotein cholesterol and homocysteine, etc. Computed tomography scans were performed to collect neuroimaging data, including hematoma location, volume and presence of intraventricular hemorrhage. Univariate and multivariate linear analyses were adopted to explore the relationship between clinical and laboratory variables and bilateral PD respectively. RESULTS: In ICH patients, PD was lower (indicating disturbed autoregulation) both on the ipsilateral (37.53±17.78 degree, P<0.001) and contralateral (34.45±14.92 degree, P<0.001) side of hematoma compared with healthy controls (56.13±16.11 degree). Hematoma volume was independently associated with ipsilateral PD according to multivariate analysis (ß=-0.383, P=0.024) after adjustment for clinical and laboratory factors. CONCLUSIONS: DCA is bilaterally disturbed after supratentorial ICH. Larger hematoma volume is likely to independently predict poorer cerebral autoregulation status ipsilateral to hematoma.