A deep learning approach to dysphagia-aspiration detecting algorithm through pre- and post-swallowing voice changes.

Introduction: This study aimed to identify differences in voice characteristics and changes between patients with dysphagia-aspiration and healthy individuals using a deep learning model, with a focus on under-researched areas of pre- and post-swallowing voice changes in patients with dysphagia. We hypothesized that these variations may be due to weakened muscles and blocked airways in patients with dysphagia. Methods: A prospective cohort study was conducted on 198 participants aged >40 years at the Seoul National University Bundang Hospital from October 2021 to February 2023. Pre- and post-swallowing voice data of the participants were converted to a 64-kbps mp3 format, and all voice data were trimmed to a length of 2 s. The data were divided for 10-fold cross-validation and stored in HDF5 format with anonymized IDs and labels for the normal and aspiration groups. During preprocessing, the data were converted to Mel spectrograms, and the EfficientAT model was modified using the final layer of MobileNetV3 to effectively detect voice changes and analyze pre- and post-swallowing voices. This enabled the model to probabilistically categorize new patient voices as normal or aspirated. Results: In a study of the machine-learning model for aspiration detection, area under the receiver operating characteristic curve (AUC) values were analyzed across sexes under different configurations. The average AUC values for males ranged from 0.8117 to 0.8319, with the best performance achieved at a learning rate of 3.00e-5 and a batch size of 16. The average AUC values for females improved from 0.6975 to 0.7331, with the best performance observed at a learning rate of 5.00e-5 and a batch size of 32. As there were fewer female participants, a combined model was developed to maintain the sex balance. In the combined model, the average AUC values ranged from 0.7746 to 0.7997, and optimal performance was achieved at a learning rate of 3.00e-5 and a batch size of 16. Conclusion: This study evaluated a voice analysis-based program to detect pre- and post-swallowing changes in patients with dysphagia, potentially aiding in real-time monitoring. Such a system can provide healthcare professionals with daily insights into the conditions of patients, allowing for personalized interventions. Clinical Trial Registration: ClinicalTrials.gov, identifier NCT05149976.

Comparison of the rate of concomitant proximal venous stenosis between the upper and lower extremities in patients with secondary lymphedema undergoing lymphaticovenous anastomosis.

BACKGROUND: Concomitant iatrogenic proximal venous stenosis increases venous pressure and can be a risk factor for unfavorable outcomes of lymphaticovenular anastomosis (LVA) in extremities with secondary lymphedema. This study investigated the frequency and relevant factors of venous stenosis in patients diagnosed with secondary lymphedema who underwent LVA. METHODS: Patients who underwent preoperative computed tomographic venography (CTV) and LVA for secondary lymphedema of the extremities from October 2018 to March 2022 were included. The incidence of proximal venous stenosis in the affected limb on preoperative CTV and the rate of endovascular intervention were compared between upper and lower extremities. Factors affecting proximal venous stenosis were identified through multivariable analysis using independent variables, including patient age, body mass index, comorbidities, smoking history, radiation therapy, duration of lymphedema, and location of lymphedema. RESULTS: A total of 211 patients were analyzed, including 83 patients with upper extremity and 128 patients with lower extremity lymphedema. The incidence of proximal venous stenosis in the preoperative CTV was 32.5% and 7.8% in upper extremity, and lower extremity lymphedema, respectively (P < .001). The incidence of venous stenosis requiring endovascular intervention was significantly higher in the upper extremity compared with the lower extremity (16.9% vs 6.3%; P = .014). In multivariable analysis, risk factors affecting incidence of venous stenosis requiring endovascular intervention was the patient age (P = .007) and upper extremity (P = .009). CONCLUSIONS: Preoperative evaluation and treatment of venous stenosis in extremities with secondary lymphedema are necessary before LVA surgery, particularly in upper extremity lymphedema.

AP collagen peptides (APCPs) promote hair growth by activating the GSK-3ß/ß-catenin pathway and improve hair condition.

AP collagen peptides (APCPs) are enzymatically decomposed collagen peptides that contain tri-peptides such as glycine-proline-hydroxyproline. We found that APCPs increased the proliferation of both human dermal papilla cells (hDPCs) and human outer root sheath cells (hORSCs). APCPs also stimulated the secretion of several growth factors, including IGFBP-6, PDGF-AB, PIGF and VEGF in hDPCs. Moreover, APCPs enhanced the phosphorylation of Akt(Ser473), GSK-3ß(Ser9) and ß-catenin(Ser675), indicating the activation of the GSK-3ß/ß-catenin signalling pathway. Ex vivo culture of human hair follicles (hHFs) tissue and in vivo patch assay revealed that APCPs promoted the elongation of hHFs and the induction of new hair shafts. In a mouse model, APCPs significantly promoted the transition from telogen to anagen phase and prolonged anagen phase, resulting in increased hair growth. APCPs also improved the thickness, amino acid content (cystine and methionine) and roughness of mouse hair. Taken together, these findings demonstrate that APCPs accelerate hair growth and contribute to overall hair health. Therefore, APCPs have the potential to be utilized as a food supplement and ingredient for preventing hair loss and maintaining hair health.

A multicenter, double-blind, randomized, parallel-group, active-controlled, phase 3 clinical trial to compare the effectiveness and safety of two botulinum toxin type A formulations for improving moderate to severe glabellar wrinkles in Asians.

Botulinum toxin type A (BoNT-A) was first isolated in 1946, and since then, several formulations have been developed and widely used to treat wrinkles by inducing muscle paralysis. This multicenter, double-blind, randomized, parallel-group, active-controlled phase 3 clinical trial was designed to evaluate the efficacy and safety of a newly developed BoNT-A formulation, BMI2006, in improving moderate to severe glabellar wrinkles and to compare with existing onabotulinumtoxin A (OBoNT) injections. A total of 276 subjects were enrolled and received 20 units of the randomized material, which was intramuscularly injected into five different locations on the forehead. The primary endpoint, assessed at 4 weeks, showed no statistically significant difference in the improvement rate of glabellar wrinkles between the two groups, with BMI2006 demonstrating non-inferiority to comparator BoNT-A. Secondary endpoints, evaluated by both treating investigators and independent investigators, also exhibited similar improvement rates throughout the study period. Both groups reported high levels of satisfaction with no statistical difference between the two groups. Safety evaluations indicated mild and transient adverse events, with no serious reactions observed. In conclusion, BMI2006 is an effective and safe BoNT-A for treating glabellar wrinkles with an expected duration of action between 8 and 12 weeks.

The efficacy of enoxolone in reducing erythema and pain after laser treatment: A randomized split-face pilot study.

BACKGROUND: Enoxolone, derived from licorice, possesses potent anti-inflammatory, and antioxidant properties. However, its effectiveness in alleviating post-laser reactions has not been extensively studied. AIMS: This randomized split-face pilot study aimed to evaluate the effects of enoxolone on skin following laser treatment. PATIENTS/METHODS: Ten healthy subjects underwent non-ablative 1550 nm Er:Glass fractional laser treatment and then randomly applied a moisturizer without enoxolone on one side of the face and a dermo-cosmetic formular containing 2% enoxolone mixed with the same moisturizer on the other side. The erythema index (EI), clinician's erythema assessment (CEA), and pain scores were recorded at 30 min, 60 min, and 24 h posttreatment. RESULTS: The group treated with enoxolone showed significantly lower EI and CEA compared to the control group at 24 h posttreatment. Additionally, pain scores were notably reduced in the enoxolone-treated group 30 min after treatment. CONCLUSIONS: This study suggests that dermo-cosmetic formular containing 2% enoxolone is effective in reducing erythema and pain following laser treatment.

Efficacy and Safety of PrabotulinumtoxinA in Subjects With Benign Masseteric Hypertrophy: A Double-Blind, Randomized, Placebo-Controlled, Multicenter, Phase 3 Trial and Open-Label Extension Study.

BACKGROUND: Despite the widespread use of botulinum toxin (BTX) injection for the treatment of masseter muscle hypertrophy (MMH), there is no standard treatment option. OBJECTIVE: We report the efficacy and safety for BTX in MMH over a period of 48 weeks. METHODS: In double-blinded, placebo-controlled phase 3 trials, 180 patients (randomized 1:1) received treatment with placebo (normal saline) or prabotulinumtoxinA (48 units). Masseter muscle thickness (at maximal clenching and resting positions), 3D imaging analysis, and masseter muscle hypertrophy scale grades were analyzed at each time point. After the 24-week CORE study, all patients who met the same criteria of the CORE study at week 24 ( n = 114) received only prabotulinumtoxinA, regardless of previous treatment, for an additional 24 weeks (48 weeks in total) for the open-label extension study. RESULTS: The largest differences in mean and percent changes from baseline in masseter muscle thickness were observed at 12 weeks, and there were significant differences between the 2 groups at all time points (all p < .001). The effect was independent of the number of injections. No serious adverse event was observed. CONCLUSION: PrabotulinumtoxinA could effectively ameliorate MMH without major complications.

Prediction of dysphagia aspiration through machine learning-based analysis of patients' postprandial voices.

BACKGROUND: Conventional diagnostic methods for dysphagia have limitations such as long wait times, radiation risks, and restricted evaluation. Therefore, voice-based diagnostic and monitoring technologies are required to overcome these limitations. Based on our hypothesis regarding the impact of weakened muscle strength and the presence of aspiration on vocal characteristics, this single-center, prospective study aimed to develop a machine-learning algorithm for predicting dysphagia status (normal, and aspiration) by analyzing postprandial voice limiting intake to 3 cc. METHODS: Conducted from September 2021 to February 2023 at Seoul National University Bundang Hospital, this single center, prospective cohort study included 198 participants aged 40 or older, with 128 without suspected dysphagia and 70 with dysphagia-aspiration. Voice data from participants were collected and used to develop dysphagia prediction models using the Multi-Layer Perceptron (MLP) with MobileNet V3. Male-only, female-only, and combined models were constructed using 10-fold cross-validation. Through the inference process, we established a model capable of probabilistically categorizing a new patient's voice as either normal or indicating the possibility of aspiration. RESULTS: The pre-trained models (mn40_as and mn30_as) exhibited superior performance compared to the non-pre-trained models (mn4.0 and mn3.0). Overall, the best-performing model, mn30_as, which is a pre-trained model, demonstrated an average AUC across 10 folds as follows: combined model 0.8361 (95% CI 0.7667-0.9056; max 0.9541), male model 0.8010 (95% CI 0.6589-0.9432; max 1.000), and female model 0.7572 (95% CI 0.6578-0.8567; max 0.9779). However, for the female model, a slightly higher result was observed with the mn4.0, which scored 0.7679 (95% CI 0.6426-0.8931; max 0.9722). Additionally, the other models (pre-trained; mn40_as, non-pre-trained; mn4.0 and mn3.0) also achieved performance above 0.7 in most cases, and the highest fold-level performance for most models was approximately around 0.9. The 'mn' in model names refers to MobileNet and the following number indicates the 'width_mult' parameter. CONCLUSIONS: In this study, we used mel-spectrogram analysis and a MobileNetV3 model for predicting dysphagia aspiration. Our research highlights voice analysis potential in dysphagia screening, diagnosis, and monitoring, aiming for non-invasive safer, and more effective interventions. TRIAL REGISTRATION: This study was approved by the IRB (No. B-2109-707-303) and registered on clinicaltrials.gov (ID: NCT05149976).