AIM: To analyze whether alterations of voxel mirror homology connectivity (VMHC) values, as determined by resting-state functional magnetic resonance imaging (rs-fMRI), occur in cerebral regions of patients with hypertensive retinopathy (HR) and to determine the relationship between VMHC values and clinical characteristics in patients with HR. METHODS: Twenty-one patients with HR and 21 age-matched healthy controls (HCs) were assessed by rs-fMRI scanning. The functional connectivity between the hemispheres of the cerebrum was assessed by measuring VMHC, with the ability of VMHC to distinguish between the HR and HC groups assessed using receiver operating characteristic (ROC) curve analysis. Differences in the demographic and clinical characteristics of the HR and HC groups were analyzed by independent sample t-tests. The relationship between average VMHC in several brain areas of HR patients and clinical features was determined using Pearson correlation analysis. RESULTS: Mean VMHC values of the bilateral cuneus gyrus (BA19), bilateral middle orbitofrontal gyrus (BA47), bilateral middle temporal gyrus (BA39) and bilateral superior medial frontal gyrus (BA9) were lower in the HR than in the HC group. CONCLUSION: VMHC values can predict the development of early HR, prevent the transformation of hypertensive microangiopathy, and provide useful information explaining the changes in neural mechanism associated with HR.

AIM: To investigate changes in local brain activity after laser assisted in situ keratomileusis (LASIK) in myopia patients, and further explore whether post-LASIK (POL) patients and healthy controls (HCs) can be distinguished by differences in dynamic amplitude of low-frequency fluctuations (dALFF) in specific brain regions. METHODS: The resting-state functional magnetic resonance imaging (rs-fMRI) data were collected from 15 myopic patients who underwent LASIK and 15 matched healthy controls. This method was selected to calculate the corresponding dALFF values of each participant, to compare dALFF between the groups and to determine whether dALFF distinguishes reliably between myopic patients after LASIK and HCs using the linear support vector machine (SVM) permutation test (5000 repetitions). RESULTS: dALFF was lower in POL than in HCs at the right precentral gyrus and right insula. Classification accuracy of the SVM was 89.1% (P<0.001). CONCLUSION: The activity of spontaneous neurons in the right precentral gyrus and right insula of myopic patients change significantly after LASIK. SVM can correctly classify POL patients and HCs based on dALFF differences.

Breast cancer (BC) is caused by the uncontrolled proliferation of breast epithelial cells followed by malignant transformation, and it has the highest incidence among female malignant tumors. The metastasis of BC occurs through direct and lymphatic spread. Although ocular metastasis is relatively rare, it is a good indicator of a worse prognosis. We used machine learning (ML) to establish a model to analyze the risk factors of BC eye metastasis. The clinical data of 2225 patients with BC from 2003 to 2019 were collected and randomly classified into the training and test sets using a ratio of 7:3. Based on the presence or absence of eye metastasis, the patients with BC were classified into the ocular metastasis (OM) and non-ocular metastasis (NOM) groups. Univariate and multivariate logistic regression analyses and least absolute shrinkage and selection operator (LASSO) were conducted. We used six ML algorithms to establish a predictive BC model and used 10-fold cross-validation for internal verification. The area under the receiver operating characteristic (ROC) curve was used to evaluate the predictive ability of the model. In addition, we established a web hazard calculator depending on the best-performing model to facilitate its clinical application. Shapley additive interpretation (SHAP) was used to determine the risk factors and the interpretability of the black box model. Univariate logistic regression analysis showed that histopathology (other types), axillary lymph node metastasis (ALNM) (> 4), Ca2+, total cholesterol (TC), low-density lipoprotein (LDL), apolipoprotein A (ApoA), carcinoembryonic antigen (CEA), carbohydrate antigen (CA) 125, CA153, CA199, alkaline phosphatase (ALP), and hemoglobin (Hb) were risk factors for BC eye metastasis. Multivariate logistic regression analysis showed that CA153, ApoA, and LDL were hazardous components for BC eye metastasis. LASSO showed that ALNM, LDL, CA125, Hb, ALP, and CA199 were the first six key variables that were useful for the diagnosis of ocular metastasis in breast cancer. Bootstrapped aggregation (BAG) demonstrated the discriminative ability (area under ROC curve [AUC] = 0.992, accuracy = 0.953, sensitivity = 0.987). Based on this, we applied the BAG machine learning model to build an online web computing system to help clinicians assist in determining the risk of BC eye metastasis. In addition, two typical cases are analyzed to determine the interpretability of the model. We used ML to establish a risk prediction model for BC ocular metastasis, and BAG showed the greatest performance. The model can predict the risk of OM in patients with BC, facilitate early and timely diagnosis and treatment, and reduce the burden on society.

AIM: To study functional brain abnormalities in patients with hypertensive retinopathy (HR) and to discuss the pathophysiological mechanisms of HR by fractional amplitude of low-frequency fluctuations (fALFFs) method. METHODS: Twenty HR patients and 20 healthy controls (HCs) were respectively recruited. The age, gender, and educational background characteristics of the two groups were similar. After functional magnetic resonance imaging (fMRI) scanning, the subjects' spontaneous brain activity was evaluated with the fALFF method. Receiver operating characteristic (ROC) curve analysis was used to classify the data. Further, we used Pearson's correlation analysis to explore the relationship between fALFF values in specific brain regions and clinical behaviors in patients with HR. RESULTS: The brain areas of the HR group with lower fALFF values than HCs were the right orbital part of the middle frontal gyrus (RO-MFG) and right lingual gyrus. In contrast, the values of fALFFs in the left middle temporal gyrus (MTG), left superior temporal pole (STP), left middle frontal gyrus (MFG), left superior marginal gyrus (SMG), left superior parietal lobule (SPL), and right supplementary motor area (SMA) were higher in the HR group. The results of a t-test showed that the average values of fALFFs were statistically significantly different in the HR group and HC group (P<0.001). The fALFF values of the left middle frontal gyrus in HR patients were positively correlated with anxiety scores (r=0.9232; P<0.0001) and depression scores (r=0.9682; P<0.0001). CONCLUSION: fALFF values in multiple brain regions of HR patients are abnormal, suggesting that these brain regions in HR patients may be dysfunctional, which may help to reveal the pathophysiological mechanisms of HR.

Moyamoya disease (MMD) is a structural abnormality of the cerebral vasculature characterized by cerebral ischemia, and is rare but its incidence is increasing. Digital subtraction angiography (DSA) of the brain is the primary means of diagnosing and evaluating this disease. But its high price and invasiveness limit its use as a monitoring tool for disease progression. As a non-invasive test for ophthalmic disorders, the optical coherence tomography angiography (OCTA) is widely used. In addition to ophthalmic diseases, OCTA has also been used in some neurological diseases. The aim of this study was to assess fundus changes in patients with MMD by OCTA and to investigate whether these changes could be a diagnostic and assessment marker for MMD. This study evaluated cerebral vessels, superficial macular capillary vessel density (SMC-VD) and macular retinal thickness (MRT) in subjects in the non-operated group (nGO), operated group (OG) and healthy controls (HC) using DSA, OCTA and other techniques. Analyses of variance (ANOVA) and Bonferroni post hoc analysis were used to calculate statistical differences between the three groups. Correlations between SMC-VD and MRT were assessed using Pearson correlation analysis. In addition, the ability of the SMC-VD and the MRT to distinguish MMD from HC was analyzed using receiver operating characteristic (ROC) curves. We found that the SMC-VD and MRT in the nOG group were significantly lower than those in the HC group and had not returned to normal levels at one month postoperatively. In the nOG, the SMC-VD and MRT were positively correlated in the Full region (6*6 mm) and in the Inner region (3*3 mm), and in many subregions they showed high ability to distinguish MMD from HC. The above findings indicate significant reduction in the SMC-VD and the MRT in patients with MMD even in the absence of ocular clinical manifestation. Most importantly, SMC-VD and MRT have a strong ability to distinguish between MMD patients and HC, suggesting that OCTA, a relatively inexpensive and non-invasive method, is useful in assessing cerebrovascular changes in MMD patients.

Acute pancreatitis, a common exocrine inflammatory disease affecting the pancreas, is characterized by intense abdominal pain and multiple organ dysfunction. However, the alterations in retinal blood vessels among individuals with acute pancreatitis remain poorly understood. This study employed optical coherence tomography angiography (OCTA) to examine the superficial and deep retinal blood vessels in patients with pancreatitis. Sixteen patients diagnosed with pancreatitis (32 eyes) and 16 healthy controls (32 eyes) were recruited from the First Affiliated Hospital of Nanchang University for participation in the study. Various ophthalmic parameters, such as visual acuity, intraocular pressure, and OCTA image for retina consisting of the superficial retinal layer (SRL) and the deep retinal layer (DRL), were recorded for each eye. The study observed the superficial and deep retinal microvascular ring (MIR), macrovascular ring (MAR), and total microvessels (TMI) were observed. Changes in retinal vascular density in the macula through annular partitioning (C1-C6), hemispheric quadrant partitioning (SR, SL, IL, and IR), and early diabetic retinopathy treatment studies (ETDRS) partitioning methods (R, S, L, and I). Correlation analysis was employed to investigate the relationship between retinal capillary density and clinical indicators. Our study revealed that in the superficial retinal layer, the vascular density of TMI, MIR, MAR, SR, IR, S, C2, C3 regions were significantly decreased in patients group compared with the normal group. For the deep retinal layer, the vascular density of MIR, SR, S, C1, C2 regions also reduced in patient group. The ROC analysis demonstrated that OCTA possesses significant diagnostic performance for pancreatitis. In conclusion, patients with pancreatitis may have retinal microvascular dysfunction, and OCTA can be a valuable tool for detecting alterations in ocular microcirculation in pancreatitis patients in clinical practice.

Myocardial infarction is defined as a sudden decrease or interruption in blood flow to the coronary arteries, causing ischemic necrosis of the corresponding cardiomyocytes. It is unclear whether systemic macrovascular alterations are associated with retinal microvascular changes. This study utilized optical coherence tomography angiography (OCTA) to compare variations in conjunctival vascular density and fundus retinal vessel density between patients with myocardial infarction (MI) and healthy controls. This study recruited 16 patients (32 eyes) with MI and 16 healthy controls (32 eyes). The superficial retinal layer (SRL), deep retinal layer (DRL) and conjunctival capillary plexus in each eye were evaluated by OCTA. Parameters measured included the density of the temporal conjunctival capillary, retinal microvascular (MIR) and macrovascular (MAR) alterations and total MIR (TMI). The microvascular density of each retinal region was evaluated by the hemisphere segmentation (SR, SL, IL, and IR), annular partition (C1, C2, C3, C4, C5 and C6), and modified early treatment of diabetic retinopathy study (R, S, L, and I) methods. In the macular area, the superficial and deep retinal microvascular densities displayed notable variations. In the superficial layers, the superficial TMI, superficial MIR, and superficial MAR, as well as densities in the SL, IL, S, L, C1, C2, C5 and C6 regions, were significantly lower in MI patients (p < 0.05 each). In the deep layers, the deep MIR and deep TMI), as well as densities in the SL, IL, L, C1, C2 and C6 regions were significantly lower in MI patients (p < 0.05 each). In contrast, the conjunctival microvascular density was significantly higher in MI patients than in healthy controls (p < 0.001). The microvascular densities measured in the deep and superficial retinal layers and in the conjunctiva differ in MI patients and healthy controls. OCTA is effective in detecting changes in the ocular microcirculation.

AIM: To establish a meaningful standard for diagnosing ocular metastasis (OM) in menopausal breast cancer (BC) women, and explore the relationship between CA-153, CA-125, apolipoprotein A, and OM. METHODS: A total of 1362 menopausal female BC patients with OM volunteered to take part in this study between July 2012 and July 2022. Women with BC who are menopausal were found to have an OM incidence of 1.6%. Furthermore, CA-153, CA-125, and apolipoprotein A (Apo A) all contributed to OM in women with BC who are postmenopausal according to binary logistic regression. Receiver operating curve (ROC) analysis was used to assess the diagnostic value of OM in patients with BC. RESULTS: Both CA-153 and CA-153+CA-125 showed a higher sensitivity of 95.45%, whereas CA-153+Apo A illustrated the highest specificity of 99.02%. Moreover, CA-153 and CA-153+CA-125 had higher areas under the curve (AUC) of 0.973. CONCLUSION: The data indicate that the serum concentrations of CA-153 exhibited the most significant predictors of the diagnosis of OM in menopausal women with BC. The current study researches the utility of risk factors in predicting of OM in menopausal BC women and put forward the latest suggestions on their clinical application.

AIMS:This study aimed to investigate the resting-state functional connectivity and topologic characteristics of brain networks in patients with diabetic optic neuropathy (DON). METHODS:Resting-state functional magnetic resonance imaging scans were performed on 23 patients and 41 healthy control (HC) subjects. We used independent component analysis and graph theoretical analysis to determine the topologic characteristics of the brain and as well as functional network connectivity (FNC) and topologic properties of brain networks. RESULTS:Compared with HCs, patients with DON showed altered global characteristics. At the nodal level, the DON group had fewer nodal degrees in the thalamus and insula, and a greater number in the right rolandic operculum, right postcentral gyrus, and right superior temporal gyrus. In the internetwork comparison, DON patients showed significantly increased FNC between the left frontoparietal network (FPN-L) and ventral attention network (VAN). Additionally, in the intranetwork comparison, connectivity between the left medial superior frontal gyrus (MSFG) of the default network (DMN) and left putamen of auditory network was decreased in the DON group. CONCLUSION:DON patients altered node properties and connectivity in the DMN, auditory network, FPN-L, and VAN. These results provide evidence of the involvement of specific brain networks in the pathophysiology of DON.

Background:As an autoimmune disease, antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) often affects multiple organs, including the ocular system. This study aims to investigate differences in retinal thickness (RT) and retinal superficial vascular density (SVD) between patients with AAV and healthy controls (HCs) using optical coherence tomography angiography (OCTA). Currently, these differences are not clear. Methods:A total of 16 AAV individuals (32 eyes) and 16 HCs (32 eyes) were recruited to this cross-sectional study conducted in the First Affiliated Hospital of Nanchang University from June 2023 to September 2023. The study protocol conformed with the tenets of the Declaration of Helsinki (as revised in 2013). Each image observed by OCTA was divided into 9 regions using the Early Treatment Diabetic Retinopathy Study (ETDRS) subzones as a guide. Results:In the full layer, the RT of AAV patients was found to be significantly reduced in the inner superior (IS, P<0.001), outer superior (OS, P=0.003), inner temporal (IT, P=0.003), and outer temporal (OT, P<0.001) regions; inner RT was significantly lower in the IS (P=0.006), OS (P<0.001), inner nasal (IN, P=0.005), outer nasal (ON, P<0.001), and center (C, P=0.01) regions than that in HCs. Outer RT of AAV patients showed a reduction in the IS (P<0.001), as well as IT (P=0.008), and OT (P<0.001) regions. No statistically significant differences were seen in the different subregions in other different layers (P>0.05). Only the inner inferior (II) and outer inferior (OI) regions of SVD in AAV patients did not differ significantly from controls. All other regions showed a reduction in SVD. The details are as follows: IS (P<0.001), OS (P<0.001), IT (P=0.005), OT (P<0.001), IN (P<0.001), ON (P<0.001), and C (P=0.003). According to receiver operating characteristic (ROC) curve analysis, the full IS region [area under the curve (AUC): 0.8892, 95% confidence interval (CI): 0.8041-0.9742, P<0.001] had the highest diagnostic value for AAV-induced reduction in RT. The IS (AUC: 0.9121, 95% CI: 0.8322-0.9920, P<0.001) region was also the most sensitive to changes in SVD of AAV individuals. In addition, we found that SVD in the IN region (r=-0.4224, 95% CI: -0.6779 to -0.0757, P=0.02) as well as mean visual acuity (r=-0.3922, 95% CI: -0.6579 to -0.0397, P=0.03) of AAV patients were negatively correlated with disease duration. However, we did not find an association between SVD and RT in this study. Conclusions:The findings from OCTA indicated a reduction in RT and SVD among patients with AAV. OCTA allows for the evaluation of AAV-related ocular lesions and holds promise for monitoring of disease progression through regular evaluations.

AIM:To study functional brain abnormalities in patients with eye trauma (ET) and to discuss the pathophysiological mechanisms of ET. METHODS:Totally 31 ET patients and 31 healthy controls (HCs) were recruited. The age, gender, and educational background characteristics of the two groups were similar. After functional magnetic resonance imaging (fMRI) scanning, the subjects' spontaneous brain activity was evaluated with the functional connectivity (FC) method. Receiver operating characteristic (ROC) curve analysis was used to classify the data. Pearson's correlation analysis was used to explore the relationship between FC values in specific brain regions and clinical behaviors in patients with ET. RESULTS:Significantly increased FC between several regions was identified including the medial prefrontal cortex (MPFC) and left hippocampus formations (HF), the MPFC and left inferior parietal lobule (IPL), the left IPL and left medial temporal lobe (MTL), the left IPL and right MTL, and the right IPL and left MTL. No decreased region-to-region connectivity was detected in default mode network (DMN) sub-regions in patients with ET. Compared with HCs, ET patients exhibited significantly increased FC between several paired DMN regions, as follows: posterior cingulate cortex (PCC) and right HF (HF.R, t=2.196, P=0.032), right inferior parietal cortices (IPC.R) and left MTL (MTL.L, t=2.243, P=0.029), and right MTL (MTL.R) and HF.R (t=2.236, P=0.029). CONCLUSION:FC values in multiple brain regions of ET patients are abnormal, suggesting that these brain regions in ET patients may be dysfunctional, which may help to reveal the pathophysiological mechanisms of ET.

AIM: To evaluate the performance of three distinct large language models(LLM), including GPT-3.5, GPT-4, and PaLM2, in responding to queries within the field of ophthalmology, and to compare their performance with three different levels of medical professionals: medical undergraduates, master of medicine, and attending physicians.METHODS: A total of 100 ophthalmic multiple-choice tests, which covered ophthalmic basic knowledge, clinical knowledge, ophthalmic examination and diagnostic methods, and treatment for ocular disease, were conducted on three different kinds of LLM and three different levels of medical professionals(9 undergraduates, 6 postgraduates and 3 attending physicians), respectively. The performance of LLM was comprehensively evaluated from the aspects of mean scores, consistency and confidence of response, and it was compared with human.RESULTS: Notably, each LLM surpassed the average performance of undergraduate medical students(GPT-4:56, GPT-3.5:42, PaLM2:47, undergraduate students:40). Specifically, performance of GPT-3.5 and PaLM2 was slightly lower than those of master's students(51), while GPT-4 exhibited a performance comparable to attending physicians(62). Furthermore, GPT-4 showed significantly higher response consistency and self-confidence compared with GPT-3.5 and PaLM2.CONCLUSION: LLM represented by GPT-4 performs well in the field of ophthalmology, and the LLM model can provide clinical decision-making and teaching aids for clinicians and medical education.

Purpose: The performance of three different large language models (LLMS) (GPT-3.5, GPT-4, and PaLM2) in answering ophthalmology professional questions was evaluated and compared with that of three different professional populations (medical undergraduates, medical masters, and attending physicians). Methods: A 100-item ophthalmology single-choice test was administered to three different LLMs (GPT-3.5, GPT-4, and PaLM2) and three different professional levels (medical undergraduates, medical masters, and attending physicians), respectively. The performance of LLM was comprehensively evaluated and compared with the human group in terms of average score, stability, and confidence. Results: Each LLM outperformed undergraduates in general, with GPT-3.5 and PaLM2 being slightly below the master's level, while GPT-4 showed a level comparable to that of attending physicians. In addition, GPT-4 showed significantly higher answer stability and confidence than GPT-3.5 and PaLM2. Conclusion: Our study shows that LLM represented by GPT-4 performs better in the field of ophthalmology. With further improvements, LLM will bring unexpected benefits in medical education and clinical decision making in the near future.

Abstract Background Nasopharyngeal carcinoma (NPC) is a common cancer of the head and neck, and the eye is a common metastatic site of NPC. This study aimed to use machine learning (ML) to establish a clinical prediction model for ocular metastasis (OM) in NPC patients. Methods We retrospectively collected clinical data from 1,855 patients with NPC who were randomized to a training set and internal test set. Patients with NPC were divided into the OM group or the non-ocular metastasis (NOM) group. Independent risk factors for NPC-related hypertension risk were screened with multivariate logistic regression models. Six ML algorithms were used, including AdaBoost (AB), logistic regression (LR), random forest (RF), multilayer perceptron (MLP), bagging (BAG), and XGBoost (XGB). The training set was used to tune the model parameters to determine the final prediction model, and the test set was used to evaluate the training model. We compared the accuracy, sensitivity, area under the ROC curve, F1 score, and specificity of the models to determine the best machine-learning algorithm for predicting the probability of OM in NPC patients. In addition, a web calculator was developed to facilitate its clinical application. Results Among these six models, the AB model had the best differential diagnostic ability (F1 score = 0.773, area under the curve = 0.995, accuracy = 0.983, sensitivity = 0.833, and specificity = 0.985). Based on this model, an online web calculator was constructed to calculate the probability of OM in NPC patients to help clinicians differentially diagnose the disease. Finally, the Shapley Supplementary Interpretation library was used to screen the five most important risk factors for OM in NPC patients: TG, Cyfra 21 1, CA199, Hb, TC, and Pathology type. Conclusion We developed a risk prediction model for OM in NPC patients using ML methods and demonstrated that the AB model performed best among six ML models. This prediction model can help to identify patients with OM from NPC, provide early and individualized diagnosis and treatment plans, protect patients from OM from NPC, and minimize the burden on society.

In recent years, pre-trained large language models (LLMs) have achieved tremendous success in the field of Natural Language Processing (NLP). Prior studies have primarily focused on general and generic domains, with relatively less research on specialized LLMs in the medical field. The specialization and high accuracy requirements for diagnosis in the medical field, as well as the challenges in collecting large-scale data, have constrained the application and development of LLMs in medical scenarios. In the field of ophthalmology, clinical diagnosis mainly relies on doctors' interpretation of reports and making diagnostic decisions. In order to take advantage of LLMs to provide decision support for doctors, we collected three modalities of ophthalmic report data and fine-tuned the LLaMA2 model, successfully constructing an LLM termed the "Ophtha-LLaMA2" specifically tailored for ophthalmic disease diagnosis. Inference test results show that even with a smaller fine-tuning dataset, Ophtha-LLaMA2 performs significantly better in ophthalmic diagnosis compared to other LLMs. It demonstrates that the Ophtha-LLaMA2 exhibits satisfying accuracy and efficiency in ophthalmic disease diagnosis, making it a valuable tool for ophthalmologists to provide improved diagnostic support for patients. This research provides a useful reference for the application of LLMs in the field of ophthalmology, while showcasing the immense potential and prospects in this domain.

IMPORTANCE The response effectiveness of different large language models (LLMs) and various individuals, including medical students, graduate students, and practicing physicians, in pediatric ophthalmology consultations, has not been clearly established yet. OBJECTIVE Design a 100-question exam based on pediatric ophthalmology to evaluate the performance of LLMs in highly specialized scenarios and compare them with the performance of medical students and physicians at different levels. DESIGN, SETTING, AND PARTICIPANTS This survey study assessed three LLMs, namely ChatGPT (GPT-3.5), GPT-4, and PaLM2, were assessed alongside three human cohorts: medical students, postgraduate students, and attending physicians, in their ability to answer questions related to pediatric ophthalmology. It was conducted by administering questionnaires in the form of test papers through the LLM network interface, with the valuable participation of volunteers. MAIN OUTCOMES AND MEASURES Mean scores of LLM and humans on 100 multiple-choice questions, as well as the answer stability, correlation, and response confidence of each LLM. RESULTS GPT-4 performed comparably to attending physicians, while ChatGPT (GPT-3.5) and PaLM2 outperformed medical students but slightly trailed behind postgraduate students. Furthermore, GPT-4 exhibited greater stability and confidence when responding to inquiries compared to ChatGPT (GPT-3.5) and PaLM2. CONCLUSIONS AND RELEVANCE Our results underscore the potential for LLMs to provide medical assistance in pediatric ophthalmology and suggest significant capacity to guide the education of medical students.

Objective: To investigate the changes of amplitude of low-frequency fluctuation (ALFF) in brain regions of patients with hypertensive retinopathy by using resting-state functional magnetic resonance imaging (rs-fMRI) and change in the relationship of ALFF value with potential emotional and psychological changes. Methods: Thirty-one patients with hypertensive retinopathy (HR) (16 men and 15 women) and 31 healthy controls (HCs; 16 men and 15 women) matched for age, sex, and weight were enrolled in the research. The changes in mean ALFF values could reflect brain activity between HR patients and HCs. We used the independent samples t-test to evaluate different demographic and general information between the two groups. Two-sample t-test was used to detect differences of mean ALFF values in the brain region between the two groups using the same software. Results: The ALFF values in the brain areas of HR and HCs were different. HR patients had lower ALFF value in the left medial superior frontal gyrus and left middle frontal gyrus than the HCs. The higher ALFF values were found in the cerebellum (left inferior and right superior lobes, vermis) and left inferior temporal gyrus of the HR patients than the controls. Conclusion: Our findings showed fluctuations in ALFF values in the HR patients' brain regions. ALFF values reflect over or reduced activity in brain regions. Abnormal ALFF values in these brain areas can predict early HR development, preventing the malignant transformation of hypertensive microangiopathy.

Thin-section magnetic resonance imaging (MRI) can provide higher resolution anatomical structures and more precise clinical information than thick-section images. However, thin-section MRI is not always available due to the imaging cost issue. In multicenter retrospective studies, a large number of data are often in thick-section manner with different section thickness. The lack of thin-section data and the difference in section thickness bring considerable difficulties in the study based on the image big data. In this article, we introduce DeepVolume, a two-step deep learning architecture to address the challenge of accurate thin-section MR image reconstruction. The first stage is the brain structure-aware network, in which the thick-section MR images in axial and sagittal planes are fused by a multitask 3-D U-net with prior knowledge of brain volume segmentation, which encourages the reconstruction result to have correct brain structure. The second stage is the spatial connection-aware network, in which the preliminary reconstruction results are adjusted slice-by-slice by a recurrent convolutional network embedding convolutional long short-term memory (LSTM) block, which enhances the precision of the reconstruction by utilizing the previously unassessed sagittal information. We used 305 paired brain MRI samples with thickness of 1.0 mm and 6.5 mm in this article. Extensive experiments illustrate that DeepVolume can produce the state-of-the-art reconstruction results by embedding more anatomical knowledge. Furthermore, considering DeepVolume as an intermediate step, the practical and clinical value of our method is validated by applying the brain volume estimation and voxel-based morphometry. The results show that DeepVolume can provide much more reliable brain volume estimation in the normalized space based on the thick-section MR images compared with the traditional solutions.