Deep Learning Model Using Stool Pictures for Predicting Endoscopic Mucosal Inflammation in Patients With Ulcerative Colitis.

INTRODUCTION: Stool characteristics may change depending on the endoscopic activity of ulcerative colitis (UC). We developed a deep learning model using stool photographs of patients with UC (DLSUC) to predict endoscopic mucosal inflammation. METHODS: This was a prospective multicenter study conducted in 6 tertiary referral hospitals. Patients scheduled to undergo endoscopy for mucosal inflammation monitoring were asked to take photographs of their stool using smartphones within 1 week before the day of endoscopy. DLSUC was developed using 2,161 stool pictures from 306 patients and tested on 1,047 stool images from 126 patients. The UC endoscopic index of severity was used to define endoscopic activity. The performance of DLSUC in endoscopic activity prediction was compared with that of fecal calprotectin (Fcal). RESULTS: The area under the receiver operating characteristic curve (AUC) of DLSUC for predicting endoscopic activity was 0.801 (95% confidence interval [CI] 0.717-0.873), which was not statistically different from the AUC of Fcal (0.837 [95% CI, 0.767-0.899, DeLong P = 0.458]). When rectal-sparing cases (23/126, 18.2%) were excluded, the AUC of DLSUC increased to 0.849 (95% CI, 0.760-0.919). The accuracy, sensitivity, and specificity of DLSUC in predicting endoscopic activity were 0.746, 0.662, and 0.877 in all patients and 0.845, 0.745, and 0.958 in patients without rectal sparing, respectively. Active patients classified by DLSUC were more likely to experience disease relapse during a median 8-month follow-up (log-rank test, P = 0.002). DISCUSSION: DLSUC demonstrated a good discriminating power similar to that of Fcal in predicting endoscopic activity with improved accuracy in patients without rectal sparing. This study implies that stool photographs are a useful monitoring tool for typical UC.

Complications and Technical Consideration of Ultrasound-Guided Rectus Sheath Blocks: A Retrospective Analysis of 4033 Patients.

BACKGROUND: Although the ultrasound-guided rectus sheath block (RSB) is usually regarded as an easy and safe procedure in clinical settings, there is currently no report on complications incidence. Therefore, the present study investigated complications in a large cohort and described the technical considerations to minimize complications of real-time ultrasound-guided RSBs. METHODS: This was a retrospective cohort study of patients who underwent real-time ultrasound-guided RSBs for perioperative pain control in laparoscopic surgery with an umbilical port between February 1, 2017, and February 28, 2021, at the Asan Medical Center in South Korea. All RSBs were performed bilaterally using a 23-gauge Quincke needle, and a bilateral 2-block placement was regarded as 1 RSB. Patient data, including demographics, preoperative laboratory data, preoperative antiplatelet or anticoagulant medication with the duration of discontinuation, and type of surgery, were collected to show the study population characteristics and explore potential factors associated with adverse events such as hematoma. Ultrasound images of patients and adverse events of RSBs, including extrarectus sheath injections, vascular injuries, bowel injury, or local anesthetic systemic toxicity, were also analyzed accordingly. RESULTS: A total of 4033 procedures were analyzed. The mean body mass index of the patients was 24.1 (21.8-26.5) kg/m2. The preoperative laboratory data were within normal range in 4028 (99.9%) patients. Preoperative antiplatelets or anticoagulants were administered in 17.3% of the patients. Overall, 96 complications (2.4%) were observed. Among them, extrarectus sheath injection occurred in 88 cases (2.2%), which included preperitoneal injection (0.9%) and intraperitoneal injection (1.3%). Vascular injuries constituted 8 cases (0.2%) and all vascular injuries resulted in hematoma: 7 cases of inferior epigastric artery injury with rectus sheath hematoma and 1 of inferior mesenteric artery injury with retroperitoneal hematoma. Bowel injury or local anesthetic systemic toxicity was not reported. CONCLUSIONS: In this study of RSBs performed on 4033 patients using a 23-gauge Quincke needle in patients with low body mass index, there were 8 cases (0.2%) of vascular injury, all of which accompanied hematoma.

Tooth caries classification with quantitative light-induced fluorescence (QLF) images using convolutional neural network for permanent teeth in vivo.

BACKGROUND: Owing to the remarkable advancements of artificial intelligence (AI) applications, AI-based detection of dental caries is continuously improving. We evaluated the efficacy of the detection of dental caries with quantitative light-induced fluorescence (QLF) images using a convolutional neural network (CNN) model. METHODS: Overall, 2814 QLF intraoral images were obtained from 606 participants at a dental clinic using Qraypen C® (QC, AIOBIO, Seoul, Republic of Korea) from October 2020 to October 2022. These images included all the types of permanent teeth of which surfaces were smooth or occlusal. Dataset were randomly assigned to the training (56.0%), validation (14.0%), and test (30.0%) subsets of the dataset for caries classification. Moreover, masked images for teeth area were manually prepared to evaluate the segmentation efficacy. To compare diagnostic performance for caries classification according to the types of teeth, the dataset was further classified into the premolar (1,143 images) and molar (1,441 images) groups. As the CNN model, Xception was applied. RESULTS: Using the original QLF images, the performance of the classification algorithm was relatively good showing 83.2% of accuracy, 85.6% of precision, and 86.9% of sensitivity. After applying the segmentation process for the tooth area, all the performance indics including 85.6% of accuracy, 88.9% of precision, and 86.9% of sensitivity were improved. However, the performance indices of each type of teeth (both premolar and molar) were similar to those for all teeth. CONCLUSION: The application of AI to QLF images for caries classification demonstrated a good performance regardless of teeth type among posterior teeth. Additionally, tooth area segmentation through background elimination from QLF images exhibited a better performance.

Caries detection with tooth surface segmentation on intraoral photographic images using deep learning.

BACKGROUND: Intraoral photographic images are helpful in the clinical diagnosis of caries. Moreover, the application of artificial intelligence to these images has been attempted consistently. This study aimed to evaluate a deep learning algorithm for caries detection through the segmentation of the tooth surface using these images. METHODS: In this prospective study, 2348 in-house intraoral photographic images were collected from 445 participants using a professional intraoral camera at a dental clinic in a university medical centre from October 2020 to December 2021. Images were randomly assigned to training (1638), validation (410), and test (300) datasets. For image segmentation of the tooth surface, classification, and localisation of caries, convolutional neural networks (CNN), namely U-Net, ResNet-18, and Faster R-CNN, were applied. RESULTS: For the classification algorithm for caries images, the accuracy and area under the receiver operating characteristic curve were improved to 0.813 and 0.837 from 0.758 to 0.731, respectively, through segmentation of the tooth surface using CNN. Localisation algorithm for carious lesions after segmentation of the tooth area also showed improved performance. For example, sensitivity and average precision improved from 0.890 to 0.889 to 0.865 and 0.868, respectively. CONCLUSION: The deep learning model with segmentation of the tooth surface is promising for caries detection on photographic images from an intraoral camera. This may be an aided diagnostic method for caries with the advantages of being time and cost-saving.

Atrioventricular conduction disturbances immediately after hepatic graft reperfusion and their outcomes in patients undergoing liver transplantation.

Hemodynamic perturbation during hepatic graft reperfusion in patients undergoing liver transplantation (LT) is challenging and is frequently accompanied by bradyarrhythmia and even asystole. However, detailed data on electrocardiographic (ECG) changes during reperfusion are almost nonexistent, although the correct diagnosis by record is important for the treatment. We aimed to identify ECG rhythm disturbances during graft reperfusion and to investigate risk factors and outcomes. Data from 1065 consecutive patients who underwent adult LT were analyzed. The incidence, type, and detailed characteristics of ECG changes immediately after graft reperfusion were assessed using an electronically archived intraoperative ECG database. We analyzed risk factors, postoperative outcomes including major cardiovascular events, 30-day and 1-year mortalities of recipients based on the occurrence of atrioventricular (AV) block, and asystole during reperfusion. The typical pattern of postreperfusion bradyarrhythmia was progressive PR interval prolongation until a Mobitz type 1 AV block occurred. The overall incidence of AV block was 5.0% (53/1065), and 30.2% of them (16/53) had initiated as AV block and then progressed into ventricular asystole. Fulminant hepatic failure was a significant predictor for occurrence of AV block (odds ratio [OR], 7.20; 95% confidence interval, 3.38-15.32; P < 0.001). Patients with AV block showed significantly higher incidence of postoperative major cardiovascular events (P < 0.001) and 30-day mortality (P = 0.002) than those without AV block, whereas the 1-year mortality was not different between the 2 groups (P = 0.10). The postreperfusion asystole was consistently preceded by a Mobitz type 1 AV block. The occurrence of AV block and asystole appears to be an important prognosticator. Therefore, maintaining an optimal range of physiological status and gradual unclamping of the vena cava to avoid sudden atrial distension are recommended in high-risk patients during reperfusion period. Liver Transplantation 22 956-967 2016 AASLD.

A proposal of a simple epidural simulator for training novice anesthesiologists.

PURPOSE: Confirming the epidural space during epidural anesthesia relies mainly on feel and experience, which are difficult techniques for a trainee to learn. We designed an epidural simulator for trainees to experience loss of resistance (LOR) and various degrees of pressure resistance. METHODS: The simulator consists of a Perifix(®) LOR syringe and 1-, 5-, 10- and 50-mL syringes assembled by three-way stopcocks. A total of 89 anesthesiologists evaluated the simulator, given the choice of either the intermittent technique with air or continuous technique with saline. Sudden LOR and applicability of the simulator for training purposes were assessed using the numerical rating score (NRS). Pressure resistance at each lumbar structure was evaluated by the anesthesiologists using the intermittent technique with air. RESULTS: Seventy-four anesthesiologists used the intermittent technique with air and 15 used the continuous technique with saline. The NRSs for sudden LOR and the applicability for training purposes were 8 and 9 (median), respectively. The pressure resistance to a 50-mL syringe was regarded as the epidural space (odds ratio 602.3 for 5-mL syringe and 144.4 for 10-mL syringe) by 89 % of anesthesiologists using air for LOR. Resistance to the 10-mL syringe was most frequently considered as muscle, subcutaneous fat, or the interspinous ligament, while resistance to the 1-mL syringe was considered as the ligamentum flavum (odds ratio 2.3 for 5-mL syringe and 18.6 for 10-mL syringe). CONCLUSIONS: Our epidural simulator is a simple, low-cost device that can be easily constructed. It was shown to provide valid haptic feedback as a promising tool for training novice anesthesiologists.

Roadmap for providing and leveraging annotated data by cytologists in the PDAC domain as open data: support for AI-based pathology image analysis development and data utilization strategies.

Pancreatic cancer is one of the most lethal cancers worldwide, with a 5-year survival rate of less than 5%, the lowest of all cancer types. Pancreatic ductal adenocarcinoma (PDAC) is the most common and aggressive pancreatic cancer and has been classified as a health emergency in the past few decades. The histopathological diagnosis and prognosis evaluation of PDAC is time-consuming, laborious, and challenging in current clinical practice conditions. Pathological artificial intelligence (AI) research has been actively conducted lately. However, accessing medical data is challenging; the amount of open pathology data is small, and the absence of open-annotation data drawn by medical staff makes it difficult to conduct pathology AI research. Here, we provide easily accessible high-quality annotation data to address the abovementioned obstacles. Data evaluation is performed by supervised learning using a deep convolutional neural network structure to segment 11 annotated PDAC histopathological whole slide images (WSIs) drawn by medical staff directly from an open WSI dataset. We visualized the segmentation results of the histopathological images with a Dice score of 73% on the WSIs, including PDAC areas, thus identifying areas important for PDAC diagnosis and demonstrating high data quality. Additionally, pathologists assisted by AI can significantly increase their work efficiency. The pathological AI guidelines we propose are effective in developing histopathological AI for PDAC and are significant in the clinical field.

Real-time ultrasound guidance versus fluoroscopic guidance in thoracic epidural catheter placement: a single-center, non-inferiority, randomized, active-controlled trial.

INTRODUCTION: Fluoroscopy can improve the success rate of thoracic epidural catheter placement (TECP). Real-time ultrasound (US)-guided TECP was recently introduced and showed a high first-pass success rate. We tested whether real-time US-guided TECP results in a non-inferior first-pass success rate compared with that of fluoroscopy-guided TECP. METHODS: In this single-center, non-inferiority, randomized trial, the primary outcome was the comparison of the first-pass success rate of TECP between real-time US guidance (US group) and fluoroscopic guidance (fluoroscopy group). Secondary outcomes included time to identifying epidural space, procedure time, total number of needle passes, number of skin punctures, final success, and cross-over success. RESULTS: We randomly assigned 132 patients to the allocated groups. The difference in the first-pass success rate between the groups did not exceed the non-inferiority margin of 15% (US group: 66.7% vs fluoroscopy group: 68.2%; difference -1.5%, 95% exact CI: -14.9% to 11.9%). The difference in the final success rate also did not differ between the groups (98.5% vs 100.0%; difference -1.5%, 95% exact CI: -4.0% to 1.0%). The time to identifying epidural space (45.6 (34-62) vs 59.0 (42-77) s, p=0.004) and procedure time (39.5 (28-78) vs 112.5 (93-166) s, p<0.001) were significantly shorter in the US group. CONCLUSIONS: Real-time US guidance provided a non-inferior success rate and shorter time spent on preparation and procedure compared with fluoroscopic guidance in TECP. TRIAL REGISTRATION NUMBER: KCT0006521.

Comparing deep learning and handcrafted radiomics to predict chemoradiotherapy response for locally advanced cervical cancer using pretreatment MRI.

Concurrent chemoradiotherapy (CRT) is the standard treatment for locally advanced cervical cancer (LACC), but its responsiveness varies among patients. A reliable tool for predicting CRT responses is necessary for personalized cancer treatment. In this study, we constructed prediction models using handcrafted radiomics (HCR) and deep learning radiomics (DLR) based on pretreatment MRI data to predict CRT response in LACC. Furthermore, we investigated the potential improvement in prediction performance by incorporating clinical factors. A total of 252 LACC patients undergoing curative chemoradiotherapy are included. The patients are randomly divided into two independent groups for the training (167 patients) and test datasets (85 patients). Contrast-enhanced T1- and T2-weighted MR scans are obtained. For HCR analysis, 1890 imaging features are extracted and a support vector machine classifier with a five-fold cross-validation is trained on training dataset to predict CRT response and subsequently validated on test dataset. For DLR analysis, a 3-dimensional convolutional neural network was trained on training dataset and validated on test dataset. In conclusion, both HCR and DLR models could predict CRT responses in patients with LACC. The integration of clinical factors into radiomics prediction models tended to improve performance in HCR analysis. Our findings may contribute to the development of personalized treatment strategies for LACC patients.

Preemptive visceral analgesic effect of thoracic paravertebral block on postoperative opioid consumption in patients undergoing laparoscopic cholecystectomy: a prospective, randomized, assessor-blind study.

BACKGROUND: The preemptive visceral analgesic effect of regional nerve block has not been adequately investigated to date. We evaluated the preemptive visceral analgesic effect of thoracic paravertebral block (TPVB) in patients undergoing laparoscopic cholecystectomy (LC) in whom pre-incisional rectus sheath block (RSB) was used to minimize somatic surgical pain. METHODS: In this prospective, randomized, assessor-blind study, 70 patients scheduled for elective LC were randomly assigned to the pre-TPVB (n = 35) or the post-TPVB (n = 35) group. Both groups received pre-incisional RSB, and patients in the pre-TPVB group received TPVB before skin incision while those in the post-TPVB group received TPVB after skin closure. The primary outcome was the total rescue analgesic consumption (morphine equianalgesic dose) during the 24 h post-surgery. The secondary outcomes were the cumulative analgesic consumption and pain intensity for 24 h after surgery, and adverse events. RESULTS: Pre-TPVB significantly reduced total rescue analgesic consumption (estimated mean [95% CI]) during the 24 h after surgery than post-TPVB (16.9 [14.5, 19.3] vs. 25.3 [22.8, 27.7] mg, estimated difference: -8.3 [-11.8, -4.9], P < 0.001). The cumulative rescue analgesic consumption was significantly lower in the pre-TPVB group from 2-24 h after surgery (P < 0.001). The postoperative pain intensity was significantly lower in the pre-TPVB group as well at 0.5-6 h after surgery. There were no adverse events in both groups. CONCLUSIONS: Pre-incisional TPVB conferred a significant preemptive visceral analgesic effect in patients undergoing LC, and significantly reduced the amount of postoperative opioid consumption.

Exercise plus volume loading prevents orthostatic intolerance but not reduction in cerebral blood flow velocity after bed rest.

This study tested the hypothesis that reduction in cerebral blood flow (CBF) during orthostatic stress after bed rest can be ameliorated with volume loading, exercise, or both. Transcranial Doppler was used to measure changes in CBF velocity during lower body negative pressure (LBNP) before and after an 18-day bed rest in 33 healthy subjects. Subjects were assigned into four groups with similar age and sex: 1) supine cycling during bed rest (Exercise group; n = 7), 2) volume loading with Dextran infusion after bed rest to restore reduced left ventricular filling pressure (Dextran group; n = 7), 3) exercise combined with volume loading to prevent orthostatic intolerance (Ex-Dex group; n = 7), and 4) a control group (n = 12). LBNP tolerance was measured using a cumulative stress index (CSI). After bed rest, CBF velocity was reduced at a lower level of LBNP in the Control group, and the magnitude of reduction was greater in the Ex-Dex group. However, reduction in orthostatic tolerance was prevented in the Ex-Dex group. Notably, volume loading alone prevented greater reductions in CBF velocity after bed rest, but CSI was reduced still by 25%. Finally, decreases in CBF velocity during LBNP were correlated with reduction in cardiac output under all conditions (r(2) = 0.86; P = < 0.001). Taken together, these findings demonstrate that volume loading alone can ameliorate reductions in CBF during LBNP. However, the lack of associations between changes in CBF velocity and orthostatic tolerance suggests that reductions in CBF during LBNP under steady-state conditions by itself are unlikely to be a primary factor leading to orthostatic intolerance.

The analysis of nursing diagnoses determined by students for patients in rehabilitation units.

This study aimed to analyze nursing diagnoses determined by the nursing students for patients in rehabilitation unit. Data were collected from 190 case reports submitted by the nursing students who practiced in the rehabilitation unit, and analyzed on the basis of North American Nursing Diagnosis Association (NANDA) International, Inc. nursing diagnoses. Thirty different diagnoses were documented in rehabilitation unit. The most frequent nursing diagnosis was impaired physical mobility (n=68, 14.6%). The 30 diagnoses were grouped into 10 domains and 20 classes of the NANDA International, Inc. human response patterns. The average quality of nursing statements corresponded to a score of 8.63, indicating relatively good quality. The results of this study will help to improve the quality of nursing process education and provide guidelines to improve the quality of nursing care for the rehabilitation nursing situation in Korea.

Factors Associated With Rebound Pain After Patient-controlled Epidural Analgesia in Patients Undergoing Major Abdominal Surgery: A Retrospective Study.

OBJECTIVES: Although patient-controlled epidural analgesia (PCEA) is an effective form of regional analgesia for abdominal surgery, some patients experience significant rebound pain after the discontinuation of PCEA. However, risk factors for rebound pain associated with PCEA in major abdominal surgery remain unknown. This study evaluated the incidence of rebound pain related to PCEA and explored potential associated risk factors. MATERIALS AND METHODS: We performed a retrospective review of 236 patients using PCEA following hepatobiliary and pancreas surgery between 2018 and 2020 in a tertiary hospital in South Korea. Rebound pain was defined as an increase from well-controlled pain (numeric rating scale <4) during epidural analgesia to severe pain (numeric rating scale ≥7) within 24 hours of discontinuation of PCEA. Logistic regression analysis was performed to determine the factors associated with rebound pain. RESULTS: Patients were categorized into the nonrebound pain group (170 patients; 72%) and the rebound pain group (66 patients; 28%). Multivariable logistic regression analysis revealed that preoperative prognostic nutritional index below 45 (odds ratio=2.080, 95% confidential interval=1.061-4.079, P =0.033) and intraoperative transfusion (odds ratio=4.190, 95% confidential interval=1.436-12.226, P =0.009) were independently associated with rebound pain after PCEA discontinuation. DISCUSSION: Rebound pain after PCEA occurred in ~30% of patients who underwent major abdominal surgery, resulting in insufficient postoperative pain management. Preoperative low prognostic nutritional index and intraoperative transfusion may be associated with rebound pain after PCEA discontinuation.

Use of video-based telehealth services using a mobile app for workers in underserved areas during the COVID-19 pandemic: A prospective observational study.

BACKGROUND: The COVID-19 pandemic has limited face-to-face treatment, triggering a change in the structure of existing healthcare services. Unlike other groups, workers in underserved areas have relatively poor access to healthcare. OBJECTIVE: This study aimed to investigate the effects of video-based telehealth services using a mobile personal health record (PHR) app for vulnerable workers with metabolic risk factors. METHODS: A prospective observational study was conducted with 117 participants and 27 healthcare professionals for 16 weeks. Participants visited the research institution three times (at weeks 1, 8, and 16) and underwent health check-ups and used various features of the mobile PHR app. Healthcare professionals observed the participants's data using the monitoring system and performed appropriate interventions. The primary outcome measures were to evaluate the effects of services on changes in the participants' metabolic risk factors, and secondary outcome measures were to analyze changes in the participants' lifestyle and service satisfaction, and to observe service use through usage logs. One-way repeated measures ANOVA and Scheffé's test were performed to observe changes in participants' health status and lifestyle, and a paired t-test was performed to analyze changes in service satisfaction. Finally, in-depth interviews with healthcare professionals were performed using semi-structured questionnaires to understand service providers' perspectives after the end of the study. RESULTS: Systolic blood pressure (F = 7.32, P <.001), diastolic blood pressure (F = 11.30, P <.001), body weight (F = 29.53, P <.001), BMI (F = 17.31, P <.001), waist circumference (F = 17.33, P <.001), fasting blood glucose (F = 5.11, P =.007), and triglycerides (F = 4.66, P =.01) showed significant improvements with time points, whereas high-density lipoprotein cholesterol (F = 3.35, P =.067) did not. The dietary score (F = 3.26, P =.04) showed a significant improvement with time points, whereas physical activity (F = 1.06, P =.34) did not. In terms of service satisfaction, only lifestyle improvement (P <.001) showed a significant difference. COVID-19 has affected the performance of healthcare professionals, thereby changing the perspectives toward healthcare technology services. CONCLUSIONS: We evaluated the effectiveness of video-based telehealth services supporting workers' health status and lifestyle interventions using healthcare technologies such as the mobile PHR app, tele-monitoring, and video teleconsultation. Our results indicate that as a complementary means, its utility can be expanded in the field of occupational safety and health to overcome the limitations of face-to-face treatment due to COVID-19 in the future.

Comparison of sitting and prone positions for real-time ultrasound-guided thoracic epidural catheter placement: a randomized controlled trial.

INTRODUCTION: Real-time ultrasound-guided thoracic epidural catheter placement (US-TECP) has been recently introduced. Patient's position is associated with the success of spine interventions; however, the effects of position on the outcome of the procedure remain unknown. We aimed to assess the clinical usefulness of patient positioning during real-time US-TECP. METHODS: Patients were randomly assigned to the prone position group (group P) and sitting position group (group S). The primary outcome was needling time during the procedure. The secondary outcomes were time to mark space, total number of needle passes, number of skin punctures, first-pass success, final success, crossover success, and visibility of ultrasound (US) views. Global Rating Scale (GRS) score, Patient Comfort Scale score, procedural pain intensity, patient satisfaction, and procedure-related complications were also determined. RESULTS: Sixty-four patients were included in this study. The needling time was significantly shorter in group P than in group S (36.5 (26.5-51.0) vs 59.5 (34.5-152.0) s, p<0.01). The numbers of needle passes and skin punctures were significantly lesser in group P than in group S. First-pass success was higher in group P than in group S. Group P had higher GRS compared with group S. The time to mark space, final success, US visibility score, Patient Comfort Scale score, procedural pain intensity, and patient satisfaction did not differ between the groups. One patient in group S developed a vasovagal reaction. DISCUSSION: This study shows that prone position may be preferred for real-time US-TECP, considering its better clinical usefulness. TRIAL REGISTRATION NUMBER: KCT0005757.

Laboratory variables associated with low near-infrared cerebral oxygen saturation in icteric patients before liver transplantation surgery.

BACKGROUND: Although regional cerebral oxygen saturation (rSO2) measurements can detect disturbances in cerebral oxygenation, their usefulness is limited in patients with hyperbilirubinemia. We examined the relationship between rSO2 and other laboratory variables that may affect interpretation of low rSO2 in awake patients with end-stage liver disease before liver transplantation surgery. METHODS: Before induction of general anesthesia, rSO2 was measured in 164 patients with liver cirrhosis (Child class A/B/C = 19/41/104) and 8 with fulminant hepatic failure. Patients with West Haven hepatic encephalopathy of grade 3 or 4 were excluded. Relationships between rSO2 and laboratory variables were evaluated by correlation and multivariate regression, and by receiver operating characteristic curve analysis. RESULTS: Univariate analyses showed that rSO2 (median 58.5%, range 15% to 82%) correlated with serum total bilirubin, hemoglobin (Hb), creatinine, sodium, and magnesium concentrations, and prothrombin time (P < 0.001 each), but not with serum concentrations of glucose, albumin, potassium, and ammonia. Multiple logistic regression analysis showed that only elevated total bilirubin (range 0.4 to 66 mg/dL; odds ratio [OR] = 1.31; 95% confidence interval [CI] = 1.18 to 1.45) and low Hb (range 5.3 to 15.7 g/dL; OR = 0.21; 95% CI = 0.11 to 0.43) were independently related to rSO2 <50%. The optimum cutoff points for observing an rSO2 < 50% were total bilirubin >7.2 mg/dL (sensitivity 89%, specificity 90%) and Hb <9.6 g/dL (sensitivity 70%, specificity 82%). CONCLUSIONS: High total bilirubin and low Hb concentrations were independently associated with rSO2 values below 50% in end-stage liver disease patients awaiting liver transplantation. The results of this study identify patients in whom a low rSO2 may be an artifact rather than cerebral ischemia.

Quantitative gait analysis of idiopathic normal pressure hydrocephalus using deep learning algorithms on monocular videos.

A vision-based gait analysis method using monocular videos was proposed to estimate temporo-spatial gait parameters by leveraging deep learning algorithms. This study aimed to validate vision-based gait analysis using GAITRite as the reference system and analyze relationships between Frontal Assessment Battery (FAB) scores and gait variability measured by vision-based gait analysis in idiopathic normal pressure hydrocephalus (INPH) patients. Gait data from 46 patients were simultaneously collected from the vision-based system utilizing deep learning algorithms and the GAITRite system. There was a strong correlation in 11 gait parameters between our vision-based gait analysis method and the GAITRite gait analysis system. Our results also demonstrated excellent agreement between the two measurement systems for all parameters except stride time variability after the cerebrospinal fluid tap test. Our data showed that stride time and stride length variability measured by the vision-based gait analysis system were correlated with FAB scores. Vision-based gait analysis utilizing deep learning algorithms can provide comparable data to GAITRite when assessing gait dysfunction in INPH. Frontal lobe functions may be associated with gait variability measurements using vision-based gait analysis for INPH patients.

Programmed Intermittent Epidural Bolus versus Continuous Epidural Infusion in Major Upper Abdominal Surgery: A Retrospective Comparative Study.

Although recent evidence shows that the programmed intermittent epidural bolus can provide improved analgesia compared to continuous epidural infusion during labor, its usefulness in major upper abdominal surgery remains unclear. We evaluated the effect of programmed intermittent epidural bolus versus continuous epidural infusion on the consumption of postoperative rescue opioids, pain intensity, and consumption of local anesthetic by retrospective analysis of data of patients who underwent major upper abdominal surgery under ultrasound-assisted thoracic epidural analgesia between July 2018 and October 2020. The primary outcome was total opioid consumption up to 72 h after surgery. The data of postoperative pain scores, epidural local anesthetic consumption, and adverse events from 193 patients were analyzed (continuous epidural infusion: n = 124, programmed intermittent epidural bolus: n = 69). There was no significant difference in the rescue opioid consumption in the 72 h postoperative period between the groups (33.3 mg [20.0-43.3] vs. 28.3 mg [18.3-43.3], p = 0.375). There were also no significant differences in the pain scores, epidural local anesthetic consumption, and incidence of adverse events. Our findings suggest that the quality of postoperative analgesia and safety following major upper abdominal surgery were comparable between the groups. However, the use of programmed intermittent epidural bolus requires further evaluation.

Real-time ultrasound-guided low thoracic epidural catheter placement: technical consideration and fluoroscopic evaluation.

BACKGROUND AND OBJECTIVE: Thoracic epidural analgesia can significantly reduce acute postoperative pain. However, thoracic epidural catheter placement is challenging. Although real-time ultrasound (US)-guided thoracic epidural catheter placement has been recently introduced, data regarding the accuracy and technical description are limited. Therefore, this prospective observational study aimed to assess the success rate and describe the technical considerations of real-time US-guided low thoracic epidural catheter placement. METHODS: 38 patients in the prone position were prospectively studied. After the target interlaminar space between T9 and T12 was identified, the needle was advanced under real-time US guidance and was stopped just short of the posterior complex. Further advancement of the needle was accomplished without US guidance using loss-of-resistance techniques to normal saline until the epidural space was accessed. Procedure-related variables such as time to mark space, needling time, number of needle passes, number of skin punctures, and the first-pass success rate were measured. The primary outcome was the success rate of real-time US-guided thoracic epidural catheter placement, which was evaluated using fluoroscopy. In addition, the position of the catheter, contrast dispersion, and complications were evaluated. RESULTS: This study included 38 patients. The T10-T11 interlaminar space was the most location for epidural access. During the procedure, the mean time for marking the overlying skin for the procedure was 49.5±13.8 s and the median needling time was 49 s. The median number of needle passes was 1.0 (1.0-1.0). All patients underwent one skin puncture for the procedure. The first-pass and second-pass success rates were 76.3% and 18.4%, respectively. Fluoroscopic evaluation revealed that the catheter tips were all positioned in the epidural space and were usually located between T9 and T10 (84.2%). The cranial and caudal contrast dispersion were observed up to 5.4±1.6 and 2.6±1.0 vertebral body levels, respectively. No procedure-related complications occurred. CONCLUSION: Real-time US guidance appears to be a feasible option for facilitating thoracic epidural insertion. Whether or not this technique improves the procedural success and quality compared with landmark-based techniques will require additional study. TRIAL REGISTRATION NUMBER: NCT03890640.

A Worker-Centered Personal Health Record App for Workplace Health Promotion Using National Health Care Data Sets: Design and Development Study.

BACKGROUND: Personal health record (PHR) technology can be used to support workplace health promotion, and prevent social and economic losses related to workers' health management. PHR services can not only ensure interoperability, security, privacy, and data quality, but also consider the user's perspective in their design. OBJECTIVE: Using Fast Healthcare Interoperability Resources (FHIR) and national health care data sets, this study aimed to design and develop an app for providing worker-centered, interconnected PHR services. METHODS: This study considered the user's perspective, using the human-centered design (HCD) methodology, to develop a PHR app suitable for occupational health. We developed a prototype after analyzing quantitative and qualitative data collected from workers and a health care professional group, after which we performed a usability evaluation. We structured workers' PHR items based on the analyzed data, and ensured structural and semantic interoperability using FHIR, Systematized Nomenclature of Medicine-Clinical Terms (SNOMED-CT), and Logical Observation Identifiers Names and Codes (LOINC). This study integrated workers' health information scattered across different Korean institutions through an interface method, and workers' PHRs were managed through a cloud server, using Azure API for FHIR. RESULTS: In total, 562 workers from industrial parks participated in the quantitative study. The preferred data items for PHR were medication, number of steps walked, diet, blood pressure, weight, and blood glucose. The preferred features were ability to access medical checkup results, health information content provision, consultation record inquiry, and teleconsultation. The worker-centered PHR app collected data on, among others, life logs, vital signs, and medical checkup results; offered health care services such as reservation and teleconsultation; and provided occupational safety and health information through material safety data sheet search and health questionnaires. The app reflected improvements in user convenience and app usability proposed by 19 participants (7 health care professionals and 12 end users) in the usability evaluation. The After-Scenario Questionnaire (ASQ) was evaluated with a mean score of 5.90 (SD 0.34) out of 7, and the System Usability Scale (SUS) was evaluated a mean score of 88.7 (SD 4.83) out of 100. CONCLUSIONS: The worker-centered PHR app integrates workers' health information from different institutions and provides a variety of health care services from linked institutions through workers' shared PHR. This app is expected to increase workers' autonomy over their health information and support medical personnel's decision making regarding workers' health in the workplace. Particularly, the app will provide solutions for current major PHR challenges, and its design, which considers the user's perspective, satisfies the prerequisites for its utilization in occupational health.