Computers in Biology and Medicine (2022/April)
生醫工程研究所 陳冠宇 (通訊作者) / JIF:6.698 / Rank:8.85 percent (10/114)
Since nail involvement is deemed a common and important problem in patients with psoriasis, reliable repeatable specific measures are necessary to evaluate a disease severity and its response to a specific treatment. Nail Psoriasis Severity index (NAPSI) is most commonly used investigator-measured nail assessments in clinical trials. In NAPSI assessment, the nail is divided with imaginary horizontal and longitudinal lines into four quadrants. Each nail is given a score by rating the presence (1) or absence (0) of features of the four features psoriatic nail matrix disease and the four features of psoriatic nail bed disease in each of 4 quadrants, leading to a possible total score of 0–80 (for 10 fingers). To facilitate the assessment process and alleviate doctor’s workload, an automatic system for measuring nail psoriasis severity is an essential solution. In recent years, artificial intelligence has proven its strength in assisting medical diagnosis. Convolutional deep neural network (CNN) performs promisingly in object recognition and image classification. In this work, we have successfully developed a nail imaging system prototype, and developed a workflow which integrate image data collection, feature analysis, and evaluating NAPSI score. By using standardized images as training data, accuracy of model was improved, yielding a mean of accuracy for each class was 91.5 percent. This user-friendly, rapid evaluation-assisting system shows the potential of implementation in clinical setting Highlights We developed a in-house nail image acquisition sys-tem to collect standardized data. Utilized deep learning in detecting nail and distin-guishing signs of nail psoriasis. Automated NAPSI score calculation after gathering results from deep learning model.
Materials Today Bio (2022/June)
生醫工程研究所 陳冠宇 (通訊作者) / JIF:10.761 / Rank:11.22 percent (11/98)
This study introduces a highly biocompatible hybrid interface of graphene oxide (GO) and collagen type I (COL-I), whose surface topological nanostructure can effectively inhibit the differentiation of fibroblasts into myofibroblasts. The structure and roughness of this coating interface may be more easily adjusted at the nanoscale level through changes to the GO concentration, thereby effectively inducing the polarization of macrophages to the M1 state associated with an anti-fibrotic effect, without producing excessive amounts of pro-inflammatory factors. It is believe that this study makes a significant contribution to the literature because the study results demonstrate that the GO-COL interface provides definite advantages in modulating cell interactions between fibroblasts and macrophages, overcoming the limits of interfaces using GO or extra cellular matrix (ECM) alone. By adjusting the GO concentration, the level of crosslinking in the GO-COL network as well as the surface roughness can be regulated. In addition, the oxidative ability of the GO flakes on this hybrid interface provides it a unique ability to inhibit fibrosis through immune regulation, and subsequently bestows anti-fibrotic properties on materials commonly used in implants such as glass, titanium, and nitinol. Thus, these inorganic materials have great potential for use in medical implants and cell-material interfaces.
Scientific Reports (44774)
生醫所 孫家偉 / JIF:4.379 / Rank:0.232 (Q1)
Migraine is a common and complex neurovascular disorder. Clinically, the diagnosis of migraine mainly relies on scales, but the degree of pain is too subjective to be a reliable indicator. It is even more difficult to diagnose the medication-overuse headache, which can only be evaluated by whether the symptom is improved after the medication adjustment. Therefore, an objective migraine classification system to assist doctors in making a more accurate diagnosis is needed. In this research, 13 healthy subjects (HC), 9 chronic migraine subjects (CM), and 12 medication-overuse headache subjects (MOH) were measured by functional near-infrared spectroscopy (fNIRS) to observe the change of the hemoglobin in the prefrontal cortex (PFC) during the mental arithmetic task (MAT). Our model shows the sensitivity and specificity of CM are 100 percent and 75 percent, and that of MOH is 75 percent and 100 percent.The results of the classification of the three groups prove that fNIRS combines with machine learning is feasible for the migraine classification.
重要論文