Naomi Yagi

Pelvic fractures are critical injuries associated with high mortality rates, yet accurate diagnosis using conventional radiography remains challenging due to complex anatomical overlaps. This study evaluates the impact of segmentation-guided preprocessing on fracture detection from pelvic X-rays by introducing a Multi-Bone Segmentation Method. Unlike traditional approaches that restrict analysis to the pelvic ring or treat the pelvis as a single region, our framework utilizes a transformer-enhanced U-Net to distinctly isolate nine individual pelvic bones, creating an anatomically precise input for downstream classification. For segmentation, the model was trained on PXR150 and evaluated on both PXR150 and externally on AMERI PXR, while classification models were trained and evaluated separately for each dataset. We benchmarked this approach against three methods: a Conventional (raw X-ray input) baseline, a Conventional ROI-Guided Method (binary segmentation), and a Reference Segmentation (manual annotation) to establish a performance upper bound. Using the public PXR150 and private AMERI PXR datasets, we demonstrated that segmentation-guided analysis consistently enhances diagnostic precision. The Proposed Multi-Bone Method outperformed the Conventional ROI-Guided Method, achieving higher Accuracy (81.30%, 95% CI: 79.70-82.90 and 83.20%, 95% CI: 78.74-87.66) and AUROC (0.822, 95% CI: 0.794-0.851 and 0.838, 95% CI: 0.792-0.884) across both datasets. By effectively filtering non-anatomical artifacts, the multi-bone strategy achieved performance metrics close to the manually annotated reference standard. GradCAM visualizations further confirmed that multi-bone segmentation aligns model activation with clinically relevant fracture sites, enhancing interpretability. These findings underscore that anatomical completeness in segmentation is a decisive factor in improving the reliability and explainability of automated fracture detection systems.

Vacuum swallowing is a unique method for improving the pharyngeal passage of a bolus by creating subatmospheric negative pressure in the esophagus. However, whether healthy individuals and other patients with dysphagia can reproduce vacuum swallowing remains unclear. Therefore, this study aimed to assess whether healthy individuals verified using high-resolution manometry (HRM) could reproduce vacuum swallowing and evaluate its safety using a swallowing and breathing monitoring system (SBMS). Two healthy individuals who mastered vacuum swallowing taught this method to 12 healthy individuals, who performed normal and vacuum swallowing with 5 mL of water five times each. The minimum esophageal pressure and the maximum pressure of the lower esophageal sphincter (LES) were evaluated during each swallow using the HRM. Additionally, respiratory-swallowing coordination was evaluated using the SBMS. Ten individuals reproduced vacuum swallowing, and a total of 50 vacuum swallows were analyzed. The minimum esophageal pressure (-15.0 ± 4.9 vs. -46.6 ± 16.7 mmHg; P < 0.001) was significantly lower, and the maximum pressure of the LES (25.4 ± 37.7 vs. 159.5 ± 83.6 mmHg; P < 0.001) was significantly higher during vacuum swallowing. The frequencies of the I-SW and SW-I patterns in vacuum swallowing were 38.9% and 0%, respectively, using the SBMS. Vacuum swallowing could be reproduced safely in healthy participants with instruction. Therefore, instructing exhalation before and after vacuum swallowing is recommended to prevent aspiration.

PURPOSE: Swallowing dysfunction and the risk of aspiration pneumonia are frequent clinical problems in the treatment of head and neck squamous cell carcinomas (HNSCCs). Breathing-swallowing coordination is an important factor in evaluating the risk of aspiration pneumonia. To investigate breathing-swallowing discoordination after chemoradiotherapy (CRT), we monitored respiration and swallowing activity before and after CRT in patients with HNSCCs. METHODS: Non-invasive swallowing monitoring was prospectively performed in 25 patients with HNSCCs treated with CRT and grade 1 or lower radiation-induced dermatitis. Videoendoscopy, videofluoroscopy, Food Intake LEVEL Scale, and patient-reported swallowing difficulties were assessed. RESULTS: Of the 25 patients selected for this study, four dropped out due to radiation-induced dermatitis. The remaining 21 patients were analyzed using a monitoring system before and after CRT. For each of the 21 patients, 405 swallows were analyzed. Swallowing latency and pause duration after the CRT were significantly extended compared to those before the CRT. In the analysis of each swallowing pattern, swallowing immediately followed by inspiration (SW-I pattern), reflecting breathing-swallowing discoordination, was observed more frequently after CRT (p = 0.0001). In 11 patients, the SW-I pattern was observed more frequently compared to that before the CRT (p = 0.00139). One patient developed aspiration pneumonia at 12 and 23 months after the CRT. CONCLUSION: The results of this preliminary study indicate that breathing-swallowing discoordination tends to increase after CRT and could be involved in aspiration pneumonia. This non-invasive method may be useful for screening swallowing dysfunction and its potential risks.

Pelvic fractures pose significant challenges in medical diagnosis due to the complex structure of the pelvic bones. Timely diagnosis of pelvic fractures is critical to reduce complications and mortality rates. While computed tomography (CT) is highly accurate in detecting pelvic fractures, the initial diagnostic procedure usually involves pelvic X-rays (PXR). In recent years, many deep learning-based methods have been developed utilizing ImageNet-based transfer learning for diagnosing hip and pelvic fractures. However, the ImageNet dataset contains natural RGB images which are different than PXR. In this study, we proposed a two-step transfer learning approach that improved the diagnosis of pelvic fractures in PXR images. The first step involved training a deep convolutional neural network (DCNN) using synthesized PXR images derived from 3D-CT by digitally reconstructed radiographs (DRR). In the second step, the classification layers of the DCNN were fine-tuned using acquired PXR images. The performance of the proposed method was compared with the conventional ImageNet-based transfer learning method. Experimental results demonstrated that the proposed DRR-based method, using 20 synthesized PXR images for each CT, achieved superior performance with the area under the receiver operating characteristic curves (AUROCs) of 0.9327 and 0.8014 for visible and invisible fractures, respectively. The ImageNet-based method yields AUROCs of 0.8908 and 0.7308 for visible and invisible fractures, respectively.

PURPOSE: This study examined the impact of frailty and prefrailty on mid-term outcomes and rehabilitation courses after cardiac surgery. METHODS: A total of 261 patients (median age: 73 years; 30% female) who underwent elective cardiac surgery were enrolled in this study. The Japanese version of the Cardiovascular Health Study Frailty Index classified 86, 131, and 44 patients into frailty, prefrailty, and robust groups, respectively. We examined the recovery of walking ability, outcomes at discharge, mid-term all-cause mortality, and rehospitalization related to major adverse cardiovascular and cerebrovascular events (MACCE) across the three cohorts. RESULTS: The 3-year survival rates in the frailty, prefrailty, and robust groups were 87%, 97%, and 100%, respectively (p = 0.003). The free event rates of all-cause mortality and re-hospitalization related to MACCE were 59%, 79%, and 95%, respectively (p < 0.001), with a graded elevation in adjusted morbidity among patients in the prefrailty (hazard ratio [HR], 4.57; 95% confidence interval [CI], 1.08-19.4) and frailty (HR, 9.29; 95% CI 2.21-39.1) groups. Patients with frailty also experienced a delayed recovery of walking ability and a reduced number of patients with frailty were discharged home. CONCLUSION: Frailty and prefrailty adversely affect the mid-term prognosis and rehabilitation course after cardiac surgery.

Assisted reproductive technology (ART) has progressed rapidly, resulting in a great improvement in the clinical pregnancy ratio. When applying the protocol of piezo intracytoplasmic sperm injection (Piezo-ICSI), it is very important to puncture the zona pellucida and the oocyte cytoplasmic membrane without rupturing the oocyte cytoplasmic membrane. Previous studies have shown that the poor extensibility of the oocyte cytoplasmic membrane might be closely related to rupture. However, no consensus has been reached regarding how the quality of the oocyte for extensible ability or rupture possibility affects the surfaces of the oocyte on the microscopic frames. We conducted this study to provide evidence that artificial intelligence (AI) techniques are superior for predicting the tendency of oocyte rupture before puncturing on Piezo-ICSI. To inspect it, we provided a retrospective trial of 38 rupture oocytes and 55 nonruptured oocytes. This study marked the highest accuracy of 91.4% for predicting oocytes rupture using the support-vector machine method of machine learning. We conclude that AI technologies might serve an important role and provide a significant benefit to ART.

It has recently been shown that the aging population is refractory to the maintenance of swallowing function, which can seriously affect quality of life. Singing and vocal training contribute to mastication, swallowing and respiratory function. Previous studies have shown that singers have better vocal cord health. No consensus has been reached as to how vocal training affects swallowing ability. Our study was designed to establish evidence that singers are statistically superior at inducing the swallowing reflex. To test our hypothesis, we undertook a clinical trial on 55 singers and 141 non-singers (mean age: 60.1 ± 11.7 years). This cross-sectional study with propensity score matching resulted in significant differences in a repetitive saliva swallowing test among singers: 7.1 ± 2.4, n = 53 vs. non-singers: 5.9 ± 1.9, n = 53, p < 0.05. We conclude that singing can serve an important role in stabilizing the impact of voluntary swallowing on speech.

Pelvic fracture is one of the leading causes of death in the elderly, carrying a high risk of death within 1 year of fracture. This study proposes an automated method to detect pelvic fractures on 3-dimensional computed tomography (3D-CT). Deep convolutional neural networks (DCNNs) have been used for lesion detection on 2D and 3D medical images. However, training a DCNN directly using 3D images is complicated, computationally costly, and requires large amounts of training data. We propose a method that evaluates multiple, 2D, real-time object detection systems (YOLOv3 models) in parallel, in which each YOLOv3 model is trained using differently orientated 2D slab images reconstructed from 3D-CT. We assume that an appropriate reconstruction orientation would exist to optimally characterize image features of bone fractures on 3D-CT. Multiple YOLOv3 models in parallel detect 2D fracture candidates in different orientations simultaneously. The 3D fracture region is then obtained by integrating the 2D fracture candidates. The proposed method was validated in 93 subjects with bone fractures. Area under the curve (AUC) was 0.824, with 0.805 recall and 0.907 precision. The AUC with a single orientation was 0.652. This method was then applied to 112 subjects without bone fractures to evaluate over-detection. The proposed method successfully detected no bone fractures in all except 4 non-fracture subjects (96.4%).

Correct assessment of the bone healing process is required for the management of limb immobilization during the treatment of bone injuries, including fractures and defects. Although the monitoring of bone healing using ultrasound poses several advantages regarding cost and ionizing radiation exposure compared with other dominant imaging methods, such as radiography and computed tomography (CT), traditional ultrasound B-mode imaging lacks reliability and objectivity. However, the body structures can be quantitatively observed by ultrasound frequency-based methods, and therefore, the disadvantages of B-mode imaging can be overcome. In this study, we created a femoral bone hole model of a rat and observed the bone healing process using the quantitative ultrasound method and micro-CT, which provides a reliable assessment of the tissue microstructure of the bone. This study analyzed the correlation between these two assessments. The results revealed that the quantitative ultrasound measurements correlated with the CT measurements for rat bone healing. This ultrasound frequency-based method could have the potential to serve as a novel modality for quantitative monitoring of bone healing with the advantages of being less invasive and easily accessible. Impact statement Bone healing monitoring with ultrasound is advantageous as it is less invasive and easily accessible; however, the traditional B-mode method lacks reliability and objectivity. This study demonstrated that the proposed ultrasound frequency-based monitoring method can quantitatively observe bone healing and strongly correlates with the computed tomography measurements for rat bone healing. This method has the potential to become a reliable modality for monitoring bone healing.

INTRODUCTION: Postinspiratory activity, which is essential for laryngeal closure during swallowing to prevent aspiration of food into the airways, is reduced in a mouse model of tauopathy. Therefore, we hypothesized that patients at the stage of mild cognitive impairment (MCI) exhibit alterations in swallowing dynamics and coordination between swallowing and breathing. METHODS: We examined breathing-swallowing coordination in patients with MCI. Patients who scored ≥24 on the Mini-Mental State Examination and <26 on the Japanese version of the Montreal Cognitive Assessment were recruited at Sumoto Itsuki Hospital. Parameters associated with breathing-swallowing coordination were assessed using a combination of two sensors: a respiratory flow sensor and a piezoelectric sensor attached to the skin surface of the anterior neck. RESULTS: Nineteen patients met the criteria for MCI; 16 of these patients (79.5 ± 9.1 years old) scored <3 on the 10-item Eating Assessment Tool and were enrolled in the study. Their data were compared with those of an age-matched elderly cohort (79.9 ± 2.9 years old). The frequencies of swallowing during inspiration and swallowing immediately followed by inspiration in patients with MCI were 6.9% and 9.6%, respectively; these frequencies were not significantly different from those of the age-matched elderly cohort. However, the timing of swallowing in the respiratory cycle was significantly delayed in the MCI patients, and both time from the onset to the peak of laryngeal elevation and the duration between the onset of rapid laryngeal elevation and the time when the larynx returned to the resting position were significantly lengthened in this group. CONCLUSION: At the stage of MCI, breathing-swallowing coordination has already started to decline.