笠井 聡

Introduction Vietnam still faces a high burden of infectious diseases compared with developed countries, and improving its health and sanitation environment is essential for addressing both infectious and non-communicable diseases. Chest radiography is key for early detection of cardiopulmonary diseases. Artificial Intelligence (AI) research on detecting cardiopulmonary diseases from chest radiographs has advanced; however, no AI development studies have used Vietnamese data, despite its high burden of both disease types, for early detection. Therefore, we aimed to develop an AI model to classify normal and abnormal images using a Vietnamese chest radiograph dataset. Methods We retrospectively analyzed 12,827 normal and 4,644 abnormal cases from two Vietnamese institutions. Features were derived from principal component analysis and extracted using Vision Transformer and EfficientnetV2. We performed binary classification of normal and abnormal images using Light Gradient Boosting Machine with 5-fold cross-validation. Results The model achieved an F1-score of 0.668, sensitivity of 0.596, specificity of 0.931, accuracy of 0.842, and AUC of 0.897. Subgroup evaluation revealed high accuracy in both infectious and non-communicable cases, as well as in urgent cases. Conclusion We developed an AI system that classifies normal and abnormal chest radiographs with high clinical accuracy using Vietnamese data.

Fetal growth restriction, affecting up to 10% of pregnancies, is a critical factor contributing to perinatal mortality and morbidity. Ultrasound measurements of the fetal abdominal circumference (AC) are a key aspect of monitoring fetal growth. However, the routine practice of biometric obstetric ultrasounds is limited in low-resource settings due to the high cost of sonography equipment and the scarcity of trained sonographers. To address this issue, we organized the ACOUSLIC-AI (Abdominal Circumference Operator-agnostic UltraSound measurement in Low-Income Countries) challenge to investigate the feasibility of automatically estimating fetal AC from blind-sweep ultrasound scans acquired by novice operators using low-cost devices. Training data, collected from three Public Health Units (PHUs) in Sierra Leone are made publicly available. Private validation and test sets, containing data from two PHUs in Tanzania and a European hospital, are provided through the Grand-Challenge platform. All sets were annotated by experienced readers. Sixteen international teams participated in this challenge, with six teams submitting to the Final Test Phase. In this article, we present the results of the three top-performing AI models from the ACOUSLIC-AI challenge, which are publicly accessible. We evaluate their performance in fetal abdomen frame selection, segmentation, abdominal circumference measurement, and compare their performance against clinical standards for fetal AC measurement. Clinical comparisons demonstrated that the limits of agreement (LoA) for A2 in fetal AC measurements are comparable to the interobserver LoA reported in the literature. The algorithms developed as part of the ACOUSLIC-AI challenge provide a benchmark for future algorithms on the selection and segmentation of fetal abdomen frames to further minimize fetal abdominal circumference measurement variability.

Introduction Osteoporosis increases the risk of fragility fractures, especially of the lumbar spine and femur. As fractures affect life expectancy, it is crucial to detect the early stages of osteoporosis. Dual X-ray absorptiometry (DXA) is the gold standard for bone mineral density (BMD) measurement and the diagnosis of osteoporosis; however, its low screening usage is problematic. The accurate estimation of BMD using chest radiographs (CXR) could expand screening opportunities. This study aimed to indicate the clinical utility of osteoporosis screening using deep-learning-based estimation of BMD using bidirectional CXRs. Methods This study included 1,624 patients aged ≥ 20 years who underwent DXA and bidirectional (frontal and lateral) chest radiography at a medical facility. A dataset was created using BMD and bidirectional CXR images. Inception-ResNet-V2-based models were trained using three CXR input types (frontal, lateral, and bidirectional). We compared and evaluated the BMD estimation performances of the models with different input information. Results In the comparison of models, the model with bidirectional CXR showed the highest accuracy. The correlation coefficients between the model estimates and DXA measurements were 0.766 and 0.683 for the lumbar spine and femoral BMD, respectively. Osteoporosis detection based on bidirectional CXR showed higher sensitivity and specificity than the models with single-view CXR input, especially for osteoporosis based on T-score ≤ –2.5, with 92.8% sensitivity at 50.0% specificity. Discussion These results suggest that bidirectional CXR contributes to improved accuracy of BMD estimation and osteoporosis screening compared with single-view CXR. This study proposes a new approach for early detection of osteoporosis using a deep learning model with frontal and lateral CXR inputs. BMD estimation using bidirectional CXR showed improved detection performance for low bone mass and osteoporosis, and has the potential to be used as a clinical decision criterion. The proposed method shows potential for more appropriate screening decisions, suggesting its usefulness in clinical practice.

Introduction Recently, breast composition has been used as a clinical indicator for breast cancer. Although systems have been developed for objectively extracting mammary gland regions, relying on subjective judgment to identify correct mammary gland regions for breast composition can lead to significant inter-judgment variation. In this study, we automatically extracted mammary gland regions using semi-subjective corrected regions that extract only mammary gland regions while simultaneously determining quantitative regions and examining whether extracted results could be used clinically. Methods We used 670 mammograms (Pe-ru-ru, Canon Medical Systems Corporation, Tochigi, Japan). A breast physician with 30 years of experience reading mammograms subjectively evaluated mammary gland regions based on the quantitatively determined regions. We defined these images as semi-subjective corrected region images. Further, we used U-Net for segmentation and the dice coefficient as the evaluation index for the region extraction accuracy. The parameters of U-Net (number of downsampling layers, learning rate, and batch size) and the orientation of input images were changed to improve accuracy. In addition, we calculated the dice coefficient based on the breast composition type to evaluate the clinical usefulness of this study. Results The average dice coefficient with the highest accuracy was 0.882; the average dice coefficients were 0.992, 0.832, 0.904, and 0.943 for fatty, scattered, heterogeneous dense, and extremely dense regions, respectively. Conclusion The mammary gland region was automatically extracted using semi-subjective corrected region images. The average dice coefficients for the whole breast and for each breast composition were high, suggesting that this method is clinically useful.

BACKGROUND/OBJECTIVES: Research on pleasant tactile perception has primarily focused on C-tactile fibers found in hairy skin, with the forearm and face as common study sites. Recent findings of these fibers in hairless skin, such as the palms, have sparked interest in tactile stimulation on the hands. While studies have examined comfort and brain activity in passive touch, active touch remains underexplored. This study aimed to investigate differences in pleasant sensation and brain activity during active touch with stress balls of varying hardness. METHODS: Forty healthy women participated. Using functional magnetic resonance imaging (fMRI), brain activity was measured as participants alternated between gripping stress balls of soft, medium, and hard hardness and resting without a ball. Participants rated hardness and comfort on a 9-point scale. RESULTS: Soft stress balls were perceived as soft and comfortable, activating the thalamus and left insular cortex while reducing activity in the right insular cortex. Medium stress balls elicited similar perceptions and thalamic activation but with reduced right insular cortex activity. Hard stress balls caused discomfort, activating the insular cortex, thalamus, and amygdala while reducing anterior cingulate cortex activity. CONCLUSIONS: Soft stress balls may reduce aversive stimuli through perceived comfort, while hard stress balls may induce discomfort and are unlikely to alleviate stress.

Gastric radiography is an important tool for early detection of cancer. During gastric radiography, the stomach is monitored using barium and effervescent granules. However, stomach compression and physiological phenomena during the examination can cause air to escape the stomach. When the stomach contracts, physicians cannot accurately observe its condition, which may result in missed lesions. Notably, no research using artificial intelligence (AI) has explored the use of gastric radiography to estimate the amount of air in the stomach. Therefore, this study aimed to develop an AI system to estimate the amount of air inside the stomach using gastric radiographs. In this observational, cross-sectional study, we collected data from 300 cases who underwent medical screening and estimated the images with poor stomach air volume. We used pre-trained models of vision transformer (ViT) and convolutional neural network (CNN). Instead of retraining, dimensionality reduction was performed on the output features using principal component analysis, and LightGBM performed discriminative processing. The combination of ViT and CNN resulted in the highest accuracy (F-value 0.792, accuracy 0.943, sensitivity 0.738, specificity 0.978). High accuracy was maintained in the prone position, where air inside the stomach could be easily released. Combining ViT and CNN from gastric radiographs accurately identified cases of poor stomach air volume. The system was highly accurate in the prone position and proved clinically useful. The developed AI can be used to provide high-quality images to physicians and to prevent missed lesions.

PURPOSE: Identification of the molecular subtypes in breast cancer allows to optimize treatment strategies, but usually requires invasive needle biopsy. Recently, non-invasive imaging has emerged as promising means to classify them. Magnetic resonance imaging is often used for this purpose because it is three-dimensional and highly informative. Instead, only a few reports have documented the use of mammograms. Given that mammography is the first choice for breast cancer screening, using it to classify molecular subtypes would allow for early intervention on a much wider scale. Here, we aimed to evaluate the effectiveness of combining global and local mammographic features by using Vision Transformer (ViT) and Convolutional Neural Network (CNN) to classify molecular subtypes in breast cancer. METHODS: The feature values for binary classification were calculated using the ViT and EfficientnetV2 feature extractors, followed by dimensional compression via principal component analysis. LightGBM was used to perform binary classification of each molecular subtype: triple-negative, HER2-enriched, luminal A, and luminal B. RESULTS: The combination of ViT and CNN achieved higher accuracy than ViT or CNN alone. The sensitivity for triple-negative subtypes was very high (0.900, with F-value = 0.818); whereas F-value and sensitivity were 0.720 and 0.750 for HER2-enriched, 0.765 and 0.867 for luminal A, and 0.614 and 0.711 for luminal B subtypes, respectively. CONCLUSION: Features obtained from mammograms by combining ViT and CNN allow the classification of molecular subtypes with high accuracy. This approach could streamline early treatment workflows and triage, especially for poor prognosis subtypes such as triple-negative breast cancer.

INTRODUCTION: Brain magnetic resonance imaging (MRI) is important for diagnosing Alzheimer's disease (AD), and MRI acquisition time should be reduced. The current study aimed to identify which Pix2Pix-based super-resolution images can reduce errors associated with brain anatomical analysis with diffeomorphic deformation examination and MRI acquisition time. METHODS: Fifty patients with dementia who uderwent scanning using a 3-T MRI scanner in the OASIS-3 database were used to construct a super-resolution network. Network training was performed using a scaled image (64 × 64) down-sampled from the original image as the input image and paired with the original high-resolution (256 × 256) supervised image. The hippocampal volume was measured using brain anatomical analysis with diffeomorphic deformation software, which employs machine learning algorithms and performs voxel-based morphometry. Peak signal-to-noise ratio (PSNR) and Multiscale structural similarity (MS-SSIM) score were used to objectively evaluate the generated images. RESULTS: At λ = e3, the PSNR and MS-SSIM score of the generated images were 27.91 ± 1.78 dB and 0.96 ± 0.0045, respectively. This finding indicated that the generated images had the highest objective evaluation. Using the images generated at λ = e4, the left and right hippocampal volumes did not significantly differ between the original and generated super-resolution images (p = 0.76, p = 0.19, respectively). DISCUSSION: With super-resolution using the Pix2Pix network, the hippocampal volume can be accurately measured, and the MRI acquisition time can be reduced. The proposed method does not require special hardware and can be applied to previous images.

INTRODUCTION: Cardiotocography (CTG) is used to monitor and evaluate fetal health by recording the fetal heart rate (FHR) and uterine contractions (UC) over time. Among these, the detection of late deceleration (LD), the early marker of fetal mild hypoxemia, is important, and the temporal relationship between FHR and UC is an essential factor in deciphering it. However, there is a problem with UC signals generally tending to have poor signal quality due to defects in installation or obesity in pregnant women. Since obstetricians evaluate potential LD signals only from the FHR signal when the UC signal quality is poor, we hypothesized that LD could be detected by capturing the morphological features of the FHR signal using Artificial Intelligence (AI). Therefore, this study compares models using FHR only (FHR-only model) and FHR with UC (FHR + UC model) constructed using a Convolutional Neural Network (CNN) to examine whether LD could be detected using only the FHR signal. METHODS: The data used to construct the CNN model were obtained from the publicly available CTU-UHB database. We used 86 cases with LDs and 440 cases without LDs from the database, confirmed by expert obstetricians. RESULTS: The results showed high accuracy with an area under the curve (AUC) of 0.896 for the FHR-only model and 0.928 for the FHR + UC model. Furthermore, in a validation using 23 cases in which obstetricians judged that the UC signals were poor and the FHR signal had an LD-like morphology, the FHR-only model achieved an AUC of 0.867. CONCLUSION: This indicates that using only the FHR signal as input to the CNN could detect LDs and potential LDs with high accuracy. These results are expected to improve fetal outcomes by promptly alerting obstetric healthcare providers to signs of nonreassuring fetal status, even when the UC signal quality is poor, and encouraging them to monitor closely and prepare for emergency delivery.

Radiation therapy plays a crucial role in cancer treatment, necessitating precise delivery of radiation to tumors while sparing healthy tissues over multiple days. Computed tomography (CT) is integral for treatment planning, offering electron density data crucial for accurate dose calculations. However, accurately representing patient anatomy is challenging, especially in adaptive radiotherapy, where CT is not acquired daily. Magnetic resonance imaging (MRI) provides superior soft-tissue contrast. Still, it lacks electron density information, while cone beam CT (CBCT) lacks direct electron density calibration and is mainly used for patient positioning. Adopting MRI-only or CBCT-based adaptive radiotherapy eliminates the need for CT planning but presents challenges. Synthetic CT (sCT) generation techniques aim to address these challenges by using image synthesis to bridge the gap between MRI, CBCT, and CT. The SynthRAD2023 challenge was organized to compare synthetic CT generation methods using multi-center ground truth data from 1080 patients, divided into two tasks: (1) MRI-to-CT and (2) CBCT-to-CT. The evaluation included image similarity and dose-based metrics from proton and photon plans. The challenge attracted significant participation, with 617 registrations and 22/17 valid submissions for tasks 1/2. Top-performing teams achieved high structural similarity indices (≥0.87/0.90) and gamma pass rates for photon (≥98.1%/99.0%) and proton (≥97.3%/97.0%) plans. However, no significant correlation was found between image similarity metrics and dose accuracy, emphasizing the need for dose evaluation when assessing the clinical applicability of sCT. SynthRAD2023 facilitated the investigation and benchmarking of sCT generation techniques, providing insights for developing MRI-only and CBCT-based adaptive radiotherapy. It showcased the growing capacity of deep learning to produce high-quality sCT, reducing reliance on conventional CT for treatment planning.

The alignment of tissue between histopathological whole-slide-images (WSI) is crucial for research and clinical applications. Advances in computing, deep learning, and availability of large WSI datasets have revolutionised WSI analysis. Therefore, the current state-of-the-art in WSI registration is unclear. To address this, we conducted the ACROBAT challenge, based on the largest WSI registration dataset to date, including 4,212 WSIs from 1,152 breast cancer patients. The challenge objective was to align WSIs of tissue that was stained with routine diagnostic immunohistochemistry to its H&E-stained counterpart. We compare the performance of eight WSI registration algorithms, including an investigation of the impact of different WSI properties and clinical covariates. We find that conceptually distinct WSI registration methods can lead to highly accurate registration performances and identify covariates that impact performances across methods. These results provide a comparison of the performance of current WSI registration methods and guide researchers in selecting and developing methods.

Cardiotocography (CTG) is widely used to assess fetal well-being. CTG is typically obtained using ultrasound and autocorrelation methods, which extract periodicity from the signal to calculate the heart rate. However, during labor, maternal vessel pulsations can be measured, resulting in the output of the maternal heart rate (MHR). Since the autocorrelation output is displayed as fetal heart rate (FHR), there is a risk that obstetricians may mistakenly evaluate the fetal condition based on MHR, potentially overlooking the necessity for medical intervention. This study proposes a method that utilizes Doppler ultrasound (DUS) signals and artificial intelligence (AI) to determine whether the heart rate obtained by autocorrelation is of fetal origin. We developed a system to simultaneously record DUS signals and CTG and obtained data from 425 cases. The midwife annotated the DUS signals by auditory differentiation, providing data for AI, which included 30,160 data points from the fetal heart and 2160 data points from the maternal vessel. Comparing the classification accuracy of the AI model and a simple mathematical method, the AI model achieved the best performance, with an area under the curve (AUC) of 0.98. Integrating this system into fetal monitoring could provide a new indicator for evaluating CTG quality.

A comparative interpretation of mammograms has become increasingly important, and it is crucial to develop subtraction processing and registration methods for mammograms. However, nonrigid image registration has seldom been applied to subjects constructed with soft tissue only, such as mammograms. We examined whether subtraction processing for the comparative interpretation of mammograms can be performed using nonrigid image registration. As a preliminary study, we evaluated the results of subtraction processing by applying nonrigid image registration to normal mammograms, assuming a comparative interpretation between the left and right breasts. Mediolateral-oblique-view mammograms were taken from noncancer patients and divided into 1000 cases for training, 100 cases for validation, and 500 cases for testing. Nonrigid image registration was applied to align the horizontally flipped left-breast mammogram with the right one. We compared the sum of absolute differences (SAD) of the difference of bilateral images (Difference Image) with and without the application of nonrigid image registration. Statistically, the average SAD was significantly lower with the application of nonrigid image registration than without it (without: 0.0692; with: 0.0549 (p < 0.001)). In four subgroups using the breast area, breast density, compressed breast thickness, and Difference Image without nonrigid image registration, the average SAD of the Difference Image was also significantly lower with nonrigid image registration than without it (p < 0.001). Nonrigid image registration was found to be sufficiently useful in aligning bilateral mammograms, and it is expected to be an important tool in the development of a support system for the comparative interpretation of mammograms.

Recognition of mitotic figures in histologic tumor specimens is highly relevant to patient outcome assessment. This task is challenging for algorithms and human experts alike, with deterioration of algorithmic performance under shifts in image representations. Considerable covariate shifts occur when assessment is performed on different tumor types, images are acquired using different digitization devices, or specimens are produced in different laboratories. This observation motivated the inception of the 2022 challenge on MItosis Domain Generalization (MIDOG 2022). The challenge provided annotated histologic tumor images from six different domains and evaluated the algorithmic approaches for mitotic figure detection provided by nine challenge participants on ten independent domains. Ground truth for mitotic figure detection was established in two ways: a three-expert majority vote and an independent, immunohistochemistry-assisted set of labels. This work represents an overview of the challenge tasks, the algorithmic strategies employed by the participants, and potential factors contributing to their success. With an F1 score of 0.764 for the top-performing team, we summarize that domain generalization across various tumor domains is possible with today's deep learning-based recognition pipelines. However, we also found that domain characteristics not present in the training set (feline as new species, spindle cell shape as new morphology and a new scanner) led to small but significant decreases in performance. When assessed against the immunohistochemistry-assisted reference standard, all methods resulted in reduced recall scores, with only minor changes in the order of participants in the ranking.

Introduction Physical measurements of expiratory flow volume and speed can be obtained using spirometry. These measurements have been used for the diagnosis and risk assessment of chronic obstructive pulmonary disease and play a crucial role in delivering early care. However, spirometry is not performed frequently in routine clinical practice, thereby hindering the early detection of pulmonary function impairment. Chest radiographs (CXRs), though acquired frequently, are not used to measure pulmonary functional information. This study aimed to evaluate whether spirometry parameters can be estimated accurately from single frontal CXR without image findings using deep learning. Methods Forced vital capacity (FVC), forced expiratory volume in 1 s (FEV₁), and FEV₁/FVC as spirometry measurements as well as the corresponding chest radiographs of 11,837 participants were used in this study. The data were randomly allocated to the training, validation, and evaluation datasets at an 8:1:1 ratio. A deep learning network was pretrained using ImageNet. The input and output information were CXRs and spirometry test values, respectively. The training and evaluation of the deep learning network were performed separately for each parameter. The mean absolute error rate (MAPE) and Pearson’s correlation coefficient (r) were used as the evaluation indices. Results The MAPEs between the spirometry measurements and AI estimates for FVC, FEV₁ and FEV₁/FVC were 7.59% (r = 0.910), 9.06% (r = 0.879) and 5.21% (r = 0.522), respectively. A strong positive correlation was observed between the measured and predicted indices of FVC and FEV₁. The average accuracy of >90% was obtained in each estimation of spirometry indices. Bland–Altman analysis revealed good agreement between the estimated and measured values for FVC and FEV₁. Discussion Frontal CXRs contain information related to pulmonary function, and AI estimation performed using frontal CXRs without image findings could accurately estimate spirometry values. The network proposed for estimating pulmonary function in this study could serve as a recommendation for performing spirometry or as an alternative method, suggesting its utility.

Background Mammography is the modality of choice for breast cancer screening. However, some cases of breast cancer have been diagnosed through ultrasonography alone with no or benign findings on mammography (hereby referred to as non-visibles). Therefore, this study aimed to identify factors that indicate the possibility of non-visibles based on the mammary gland content ratio estimated using artificial intelligence (AI) by patient age and compressed breast thickness (CBT). Methods We used AI previously developed by us to estimate the mammary gland content ratio and quantitatively analyze 26,232 controls and 150 non-visibles. First, we evaluated divergence trends between controls and non-visibles based on the average estimated mammary gland content ratio to ensure the importance of analysis by age and CBT. Next, we evaluated the possibility that mammary gland content ratio ≥50% groups affect the divergence between controls and non-visibles to specifically identify factors that indicate the possibility of non-visibles. The images were classified into two groups for the estimated mammary gland content ratios with a threshold of 50%, and logistic regression analysis was performed between controls and non-visibles. Results The average estimated mammary gland content ratio was significantly higher in non-visibles than in controls when the overall sample, the patient age was ≥40 years and the CBT was ≥40 mm (p < 0.05). The differences in the average estimated mammary gland content ratios in the controls and non-visibles for the overall sample was 7.54%, the differences in patients aged 40–49, 50–59, and ≥60 years were 6.20%, 7.48%, and 4.78%, respectively, and the differences in those with a CBT of 40–49, 50–59, and ≥60 mm were 6.67%, 9.71%, and 16.13%, respectively. In evaluating mammary gland content ratio ≥50% groups, we also found positive correlations for non-visibles when controls were used as the baseline for the overall sample, in patients aged 40–59 years, and in those with a CBT ≥40 mm (p < 0.05). The corresponding odds ratios were ≥2.20, with a maximum value of 4.36. Conclusion The study findings highlight an estimated mammary gland content ratio of ≥50% in patients aged 40–59 years or in those with ≥40 mm CBT could be indicative factors for non-visibles.

In the past few decades, neuroscientists have studied the physiological basis of pleasant touch. Unmyelinated low-threshold mechanoreceptors are central to the study of the physiological basis of pleasant touch. Research on pleasant stimuli has mostly focused on passive stimuli, and the brain activation sites for active pleasant stimuli are not clear. Therefore, the purpose of this study was to identify brain activation sites during active pleasant stimulation of hairless skin using functional magnetic resonance imaging. Forty-two healthy subjects aged 19 years or older were asked to actively grasp in five stimulus tasks. The comfort and sensations that occurred during the tasks were investigated using a questionnaire. Significant activation was found in the middle frontal gyrus when the hair ball and slime ball were grasped, while there was significant activation in the amygdala when grasping a squeeze ball compared to the tennis ball. In a questionnaire survey of the subjects, there was a significant difference in the comfort score between the tennis ball and the squeeze ball, but no significant correlation was found between the comfort scores and the brain sites of activation. Therefore, although active stimulation with the squeeze ball significantly activated the amygdala, it was not clear that the amygdala was significantly activated by active pleasant stimulation. In the future, it will be necessary to investigate the texture of the squeeze ball in more detail, and to increase the number of subjects for further study.

INTRODUCTION: Anxiety is an emotion necessary for human survival. However, persistent and excessive anxiety can be clinically challenging. Increased anxiety affects daily life and requires early detection and intervention. Therefore, a better understanding of the neural basis of mild anxiety is needed. However, previous studies have focused primarily on resting-state functional magnetic resonance imaging (rs-fMRI) in patients with psychiatric disorders presenting with anxiety. Notably, only a few studies have been conducted on healthy participants, and the relationship between anxiety and functional brain connectivity in the healthy range remains unclear. Therefore, in this study, we aimed to clarify the differences in functional brain connectivity at different degrees of anxiety among healthy participants. METHODS: This study included 48 healthy participants with no history of psychiatric disorders. Participants were administered The General Health Questionnaire (GHQ) 60, a psychological test for assessing anxiety, and the Manifest Anxiety Scale (MAS). The participants then underwent rs-fMRI. Based on the results of each psychological test, the participants were classified into normal and anxiety groups, and the functional connectivity between the two groups was compared using a seed-to-voxel analysis. RESULTS: Comparison of functional brain connectivity between the normal and anxiety groups classified based on the GHQ60 and MAS revealed differences between brain regions comprising the salience network (SN) in both psychological tests. For the GHQ60, the anxiety group showed reduced connectivity between the right supramarginal gyrus and insular cortex compared with the normal group. However, for the MAS, the anxiety group showed reduced connectivity between the right supramarginal and anterior cingulate cortical gyri compared with the normal group. CONCLUSION: Functional connectivity within the SN was reduced in the group with higher anxiety when functional brain connectivity at different anxiety levels was examined in healthy participants. This suggests that anxiety is involved in changes in the functional brain connectivity associated with emotional processing and cognitive control.

Cardiotocography (CTG) measurements are critical for assessing fetal wellbeing during monitoring, and accurate assessment requires well-traceable CTG signals. The current FHR calculation algorithm, based on autocorrelation to Doppler ultrasound (DUS) signals, often results in periods of loss owing to its inability to differentiate signals. We hypothesized that classifying DUS signals by type could be a solution and proposed that an artificial intelligence (AI)-based approach could be used for classification. However, limited studies have incorporated the use of AI for DUS signals because of the limited data availability. Therefore, this study focused on evaluating the effectiveness of semi-supervised learning in enhancing classification accuracy, even in limited datasets, for DUS signals. Data comprising fetal heartbeat, artifacts, and two other categories were created from non-stress tests and labor DUS signals. With labeled and unlabeled data totaling 9,600 and 48,000 data points, respectively, the semi-supervised learning model consistently outperformed the supervised learning model, achieving an average classification accuracy of 80.9%. The preliminary findings indicate that applying semi-supervised learning to the development of AI models using DUS signals can achieve high generalization accuracy and reduce the effort. This approach may enhance the quality of fetal monitoring.

The early detection of glaucoma is essential in preventing visual impairment. Artificial intelligence (AI) can be used to analyze color fundus photographs (CFPs) in a cost-effective manner, making glaucoma screening more accessible. While AI models for glaucoma screening from CFPs have shown promising results in laboratory settings, their performance decreases significantly in real-world scenarios due to the presence of out-of-distribution and low-quality images. To address this issue, we propose the Artificial Intelligence for Robust Glaucoma Screening (AIROGS) challenge. This challenge includes a large dataset of around 113,000 images from about 60,000 patients and 500 different screening centers, and encourages the development of algorithms that are robust to ungradable and unexpected input data. We evaluated solutions from 14 teams in this paper and found that the best teams performed similarly to a set of 20 expert ophthalmologists and optometrists. The highest-scoring team achieved an area under the receiver operating characteristic curve of 0.99 (95% CI: 0.98-0.99) for detecting ungradable images on-the-fly. Additionally, many of the algorithms showed robust performance when tested on three other publicly available datasets. These results demonstrate the feasibility of robust AI-enabled glaucoma screening.

Background Autonomous sensory meridian response (ASMR) is a sensory response such as tingling and pleasantness from audiovisual stimuli. ASMR videos come in a wide variety of types, and personal preferences are biased. There are many reports of the effects os ASMR on sleep onset, anxiety relief, and other relaxation effects. However, prior task-oriented studies have used ASMR videos provided by the experimenter. We hypothesized that ASMR movies of a personal preference would show significantly increased activity in the nucleus accumbens, frontal cortex, and insular cortex, which are brain areas associated with relaxation. Therefore, the purpose of this study was to elucidate the neuroscientific basis for the relaxation effects of ASMR videos that match someone’s personal preferences. Methods This study included 30 healthy individuals aged ≥18 years. ASMR enthusiasts were included as the target population due to the need to have a clear preference for ASMR videos. A control video (1 type) and ASMR videos (20 types) were used as the stimulus tasks. Among the ASMR videos, those with high and low evaluation scores were considered liked and dislikedASMR videos, respectively. Functional magnetic resonance imaging was performed while the participants viewed a block design with a resting task in between. The data were analyzed using Statistical Parametric Mapping 12 to identify the areas activated by control, disliked, and liked ASMR videos. Results Emotion-related areas (the amygdala, frontal cortex, and insular cortex) not activated by control and unliked ASMR videos were activated only by liked ASMR videos. Conclusion The amygdala, frontal cortex, and insular cortex may be involved in the limbic dopamine circuits of the amygdala and middle frontal gyrus and the autonomic balance of the left and right insular cortices. This suggests the potential of positive mood and its use as a treatment for patients with anxiety and depression. These results suggest that the use of ASMR videos to match individual preferences may induce relaxation and have beneficial effects on depression and other disorders, and also support the introduction of ASMR videos in mental health care.

Recently, breast types were categorized into four types based on the Breast Imaging Reporting and Data System (BI-RADS) atlas, and evaluating them is vital in clinical practice. A Japanese guideline, called breast composition, was developed for the breast types based on BI-RADS. The guideline is characterized using a continuous value called the mammary gland content ratio calculated to determine the breast composition, therefore allowing a more objective and visual evaluation. Although a discriminative deep convolutional neural network (DCNN) has been developed conventionally to classify the breast composition, it could encounter two-step errors or more. Hence, we propose an alternative regression DCNN based on mammary gland content ratio. We used 1476 images, evaluated by an expert physician. Our regression DCNN contained four convolution layers and three fully connected layers. Consequently, we obtained a high correlation of 0.93 (p < 0.01). Furthermore, to scrutinize the effectiveness of the regression DCNN, we categorized breast composition using the estimated ratio obtained by the regression DCNN. The agreement rates are high at 84.8%, suggesting that the breast composition can be calculated using regression DCNN with high accuracy. Moreover, the occurrence of two-step errors or more is unlikely, and the proposed method can intuitively understand the estimated results.

BACKGROUND: Autonomous sensory meridian response (ASMR) is the sensation of tingling from audiovisual stimuli that leads to positive emotions. ASMR is used among young people to relax, induce sleep, reduce stress, and alleviate anxiety. However, even without experiencing tingling, ASMR is used by many young people to seek relaxation. Auditory stimulation in ASMR is thought to play the most important role among its triggers, and previous studies have used a mixture of auditory and visual stimulation and auditory stimulation. This is the first study to approach the differences between the effects of direct audiovisual and auditory stimulation from the perspective of brain function using functional magnetic resonance imaging (fMRI) and to clarify the effects of ASMR, which attracts many young people. METHODS: The subjects were 30 healthy subjects over 19 years old or older who had not experienced tingling. Brain function was imaged by fMRI while watching ASMR videos or listening to the sound files only. We administered a questionnaire based on a Likert scale to determine if the participants felt a "relaxed mood" and "tingling mood" during the task. RESULTS: Significant activation was found in the visual cortex for audiovisual stimulation and in the visual and auditory cortex for auditory stimulation. In addition, activation of characteristic sites was observed. The specific sites of activation for audiovisual stimulation were the middle frontal gyrus and the left nucleus accumbens, while the specific sites of activation for auditory stimulation were the bilateral insular cortices. The questionnaire showed no significant differences in either "relaxed mood" or "tingling mood" in response to auditory and visual stimulation or auditory stimulation alone. CONCLUSION: The results of this study showed that there was a clear difference between auditory and audiovisual stimulation in terms of the areas of activation in the brain, but the questionnaire did not reveal any difference in the subjects' mood. Audiovisual stimulation showed activation of the middle frontal gyrus and the nucleus accumbens, whereas auditory stimulation showed activation of the insular cortex. This difference in brain activation sites suggests a difference in mental health effects between auditory and audiovisual stimulation. However, future research on comparisons between those who experience tingling and those who do not, as well as investigations of physiological indices, and examination of the relationship with activated areas in the brain may show that ASMR is useful for mental health.

In this study, we compared the differences in brain activation associated with the different types of objects using functional magnetic resonance imaging (fMRI). Twenty-six participants in their 20s underwent fMRI while grasping four different types of objects. After the experiment, all of the participants completed a questionnaire based on the Likert Scale, which asked them about the sensations they experienced while grasping each object (comfort, hardness, pain, ease in grasping). We investigated the relationship between brain activity and the results of the survey; characteristic brain activity for each object was correlated with the results of the questionnaire, indicating that each object produced a different sensation response in the participants. Additionally, we observed brain activity in the primary somatosensory cortex (postcentral gyrus), the primary motor cortex (precentral gyrus), and the cerebellum exterior during the gripping task. Our study shows that gripping different objects produces activity in specific and distinct brain regions and suggests an “action appraisal” mechanism, which is considered to be the act of integrating multiple different sensory information and connecting it to actual action. To the best of our knowledge, this is the first study to observe brain activity in response to tactile stimuli and motor activity simultaneously.

Previous studies have reported a relationship between stress and brain activity, and stress has been quantitatively evaluated using near-infrared spectroscopy (NIRS). In the present study, we examined whether a relationship exists between salivary amylase levels and brain activity during the trail-making test (TMT) using mobile NIRS. This study aimed to assess stress levels by using mobile NIRS. Salivary amylase was measured with a salivary amylase monitor, and hemoglobin concentration was measured using Neu's HOT-2000. Measurements were taken four times for each subject, and the values at each measurement were evaluated. Changes in the values at the first-second, second-third, and third-fourth measurements were also analyzed. Results showed that the value of the fluctuations has a higher correlation than the comparison of point values. These results suggest that the accuracy of stress assessment by NIRS can be improved by using variability and time-series data compared with stress assessment using NIRS at a single time point.

Background: Autonomous sensory meridian response (ASMR) is used by young people to induce relaxation and sleep and to reduce stress and anxiety; it comprises somatosensation caused by audiovisual stimuli (triggers) that lead to positive emotions. Auditory stimuli play the most important role among the triggers involved in ASMR and have been reported to be more triggering than visual stimuli. On the other hand, classical music is also known to have a relaxing effect. This is the first study to clarify the difference in brain activation associated with relaxation effects between ASMR and classical music by limiting ASMR to auditory stimulation alone. Methods: Thirty healthy subjects, all over 20 years of age, underwent fMRI while listening to ASMR and classical music. We compared the differences in brain activation associated with classical music and ASMR stimulation. After the experiment, the subjects were administered a questionnaire on somatosensation and moods. After the experiment, the participants were asked whether they experienced ASMR somatosensation or frisson. They were also asked to rate the intensity of two moods during stimulation: "comfortable mood," and "tingling mood". Result: The results of the questionnaire showed that none of the participants experienced any ASMR somatosensation or frisson. Further, there was no significant difference in the ratings given to comfort mood, but there was a significant difference in those given to tingling mood. In terms of brain function, classical music and ASMR showed significant activation in common areas, while ASMR showed activation in more areas, with the medial prefrontal cortex being the main area of activation during ASMR. Conclusion: Both classical music and the ASMR auditory stimulus produced a pleasant and relaxed state, and ASMR involved more complex brain functions than classical music, especially the activation of the medial prefrontal cortex. Although ASMR was limited to auditory stimulation, the effects were similar to those of listening to classical music, suggesting that ASMR stimulation can produce a pleasant state of relaxation even if it is limited to the auditory component, without the somatic sensation of tingling. ASMR stimulation is easy to use, and appropriate for wellness purposes and a wide range of people.