Search

Background Chest pain is the most common symptom of aortic dissection (AD), but it is often confused with other prevalent cardiopulmonary diseases. We aimed to develop deep-learning models (DLMs) with ECG and CXR features to detect AD and evaluate their performance. Methods This study included 43,473 patients in the emergency department (ED) between July 2012 and December 2019 for retrospective DLM development. A development cohort including 49,071 ED records (120 AD-type A and 64 AD-type B) was used to train DLMs for ECG and CXR, and 9,904 independent ED records (40 AD-type A and 34 AD-type B) were used to validate DLM performance. Human–machine competitions of ECG and CXR were conducted. Patient characteristics and laboratory results were used to enhance the diagnostic accuracy. The DLM-enabled AD diagnostic process was prospectively evaluated in 25,885 ED visits. Results The AUCs of the ECG and CXR models were 0.918 and 0.857 for detecting AD in a human–machine competition, respectively, which were better than those of the participating physicians. In the validation cohort, the AUCs of the integrated model were 0.882, 0.960, and 0.813 in all AD, AD-type A, and AD-type B patients, respectively, with a sensitivity of 100.0% and a specificity of 81.7% for AD-type A. In patients with chest pain and D-dimer tests, the DLM could predict more precisely, achieving a positive predictive value of 62.5% in the prospective evaluation. Conclusions DLMs may serve as decision-supporting tools for AD identification and facilitate differential diagnosis in patients with acute chest pain.

BACKGROUND: Delayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for the detection of AMI, a systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis. AIMS: We aimed to develop a deep learning model (DLM) as a diagnostic support tool based on a 12-lead electrocardiogram. METHODS: This retrospective cohort study included 1,051/697 ECGs from 737/287 coronary angiogram (CAG)-validated STEMI/NSTEMI patients and 140,336 ECGs from 76,775 non-AMI patients at the emergency department. The DLM was trained and validated in 80% and 20% of these ECGs. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of the DLM. RESULTS: The AUC of the DLM for STEMI detection was 0.976 in the human-machine competition, which was significantly better than that of the best physicians. Furthermore, the DLM independently demonstrated sufficient diagnostic capacity for STEMI detection (AUC=0.997; sensitivity, 98.4%; specificity, 96.9%). Regarding NSTEMI detection, the AUC of the combined DLM and conventional cardiac troponin I (cTnI) increased to 0.978, which was better than that of either the DLM (0.877) or cTnI (0.950). CONCLUSIONS: The DLM may serve as a timely, objective and precise diagnostic decision support tool to assist emergency medical system-based networks and frontline physicians in detecting AMI and subsequently initiating reperfusion therapy.

Federated learning (FL) is a method used for training artificial intelligence models with data from multiple sources while maintaining data anonymity, thus removing many barriers to data sharing. Here we used data from 20 institutes across the globe to train a FL model, called EXAM (electronic medical record (EMR) chest X-ray AI model), that predicts the future oxygen requirements of symptomatic patients with COVID-19 using inputs of vital signs, laboratory data and chest X-rays. EXAM achieved an average area under the curve (AUC) >0.92 for predicting outcomes at 24 and 72 h from the time of initial presentation to the emergency room, and it provided 16% improvement in average AUC measured across all participating sites and an average increase in generalizability of 38% when compared with models trained at a single site using that site’s data. For prediction of mechanical ventilation treatment or death at 24 h at the largest independent test site, EXAM achieved a sensitivity of 0.950 and specificity of 0.882. In this study, FL facilitated rapid data science collaboration without data exchange and generated a model that generalized across heterogeneous, unharmonized datasets for prediction of clinical outcomes in patients with COVID-19, setting the stage for the broader use of FL in healthcare. Federated learning, a method for training artificial intelligence algorithms that protects data privacy, was used to predict future oxygen requirements of symptomatic patients with COVID-19 using data from 20 different institutes across the globe.

BACKGROUND Identifying novel colorectal cancer (CRC) prognostic biomarkers is crucial to helping clinicians make appropriate therapy decisions. Melatonin plays a major role in managing the circadian rhythm and exerts oncostatic effects on different kinds of tumours. AIM To explore the relationship between MTNR1B single-nucleotide polymorphism (SNPs) combined with gene hypermethylation and CRC prognosis. METHODS A total of 94 CRC tumour tissues were investigated. Genotyping for the four MTNR1B SNPs (rs1387153, rs2166706, rs10830963, and rs1447352) was performed using multiplex polymerase chain reaction. The relationships between the MTNR1B SNPs and CRC 5-year overall survival (OS) was assessed by calculating hazard ratios with 95%CIs. RESULTS All SNPs (rs1387153, rs2166706, rs10830963, and rs1447352) were correlated with decreased 5-year OS. In stratified analysis, rs1387153, rs10830963, and rs1447352 risk genotype combined with CDKN2A and MGMT methylation status were associated with 5-year OS. A strong cumulative effect of the four polymorphisms on CRC prognosis was observed. Four haplotypes of MTNR1B SNPs were also associated with the 5-year OS. MTNR1B SNPs combined with CDKN2A and MGMT gene methylation status could be used to predict shorter CRC survival. CONCLUSION The novel genetic biomarkers combined with epigenetic biomarkers may be predictive tool for CRC prognosis and thus could be used to individualise treatment for patients with CRC.

(1) Background: While an artificial intelligence (AI)-based, cardiologist-level, deep-learning model for detecting acute myocardial infarction (AMI), based on a 12-lead electrocardiogram (ECG), has been established to have extraordinary capabilities, its real-world performance and clinical applications are currently unknown. (2) Methods and Results: To set up an artificial intelligence-based alarm strategy (AI-S) for detecting AMI, we assembled a strategy development cohort including 25,002 visits from August 2019 to April 2020 and a prospective validation cohort including 14,296 visits from May to August 2020 at an emergency department. The components of AI-S consisted of chest pain symptoms, a 12-lead ECG, and high-sensitivity troponin I. The primary endpoint was to assess the performance of AI-S in the prospective validation cohort by evaluating F-measure, precision, and recall. The secondary endpoint was to evaluate the impact on door-to-balloon (DtoB) time before and after AI-S implementation in STEMI patients treated with primary percutaneous coronary intervention (PPCI). Patients with STEMI were alerted precisely by AI-S (F-measure = 0.932, precision of 93.2%, recall of 93.2%). Strikingly, in comparison with pre-AI-S (N = 57) and post-AI-S (N = 32) implantation in STEMI protocol, the median ECG-to-cardiac catheterization laboratory activation (EtoCCLA) time was significantly reduced from 6.0 (IQR, 5.0–8.0 min) to 4.0 min (IQR, 3.0–5.0 min) (p &lt, 0.01). The median DtoB time was shortened from 69 (IQR, 61.0–82.0 min) to 61 min (IQR, 56.8–73.2 min) (p = 0.037). (3) Conclusions: AI-S offers front-line physicians a timely and reliable diagnostic decision-support system, thereby significantly reducing EtoCCLA and DtoB time, and facilitating the PPCI process. Nevertheless, large-scale, multi-institute, prospective, or randomized control studies are necessary to further confirm its real-world performance.

Concerns regarding the high demand for skilled personnel in the science, technology, engineering, and mathematics (STEM) fields underline the importance of developing advanced information technology (IT) and programming skills among job candidates. In the past 10 years, computer programming has regained considerable attention because of rapid developments in computer programming technology. Advocates claim that computer programming cultivates other skills, including problem solving, logical thinking, and creativity. Education systems worldwide are developing courses to instruct students in programming and computational thinking. Although the importance of computer programming has been widely recognized, the systematic evaluation of the effectiveness of teaching methods and conditions that promote the learning of programming knowledge and skills has received little scholarly attention. This study thus investigated the moderating roles of learners’ construal levels and their team role ambiguity in the context of group investigation in junior high school programing courses. In this study, junior high school students were divided into pairs to develop Arduino projects. Students applied programming abilities to complete a task involving the use of Arduino boards to simulate the operation of traffic lights. Major research findings indicate that construal levels play a significant role in moderating the relationship between programming ability and learning outcome, however, role ambiguity does not significantly affect this relationship. Theoretical implications are discussed, and managerial implications are suggested.

(1) Background: Diabetic retinopathy (DR) can cause blindness. Current guidelines on diabetic eye care recommend more frequent eye examinations for more severe DR to prevent deterioration. However, close follow-up and early intervention at earlier stages are important for the prevention of disease progression of other diabetes mellitus (DM) complications. The study was designed to investigate the association between different stages of DR in type 2 DM patients and the progression of DR, (2) Methods: A total of 2623 type 2 DM patients were included in this study. In these patients, a total of 14,409 fundus color photographs was obtained. The primary outcome was the progression of DR, (3) Results: The progression of DR was highly associated with the initial grade of DR (p &lt, 0.001). Severe nonproliferative diabetic retinopathy (NPDR) was the most likely to progress to proliferative diabetic retinopathy (PDR), followed by moderate NPDR, mild NPDR, and no retinopathy. However, progression to the next stage of DR showed a different trend. We used no retinopathy as a reference. Mild NPDR showed the highest risk for progression to the next stage [hazard ratio (HR): 2.00 (95% conference interval (CI): 1.72–2.32)] relative to higher initial grades [HR (moderate NPDR): 1.82 (95% CI: 1.58–2.09) and HR (severe NPDR): 0.87 (95% CI: 0.69–1.09)]. The same trend was observed in the multivariate analysis, in which mild NPDR presented the highest risk for progression to the next stage (adjusted HR (mild NPDR): 1.95 (95% CI: 1.68–2.27), adjusted HR (moderate NPDR): 1.73 (95% CI: 1.50–1.99), and adjusted HR (severe NPDR): 0.82 (95% CI: 0.65–1.03)), (4) Conclusions: Type 2 diabetic patients with earlier-grade DR appeared to exhibit more rapid development to the next grade in our study. As these findings show, more frequent fundus color photography follow-up in earlier-grade DR patients is important to slow DR progression and awaken self-perception.

Medical records scoring is important in a health care system. Artificial intelligence (AI) with projection word embeddings has been validated in its performance disease coding tasks, which maintain the vocabulary diversity of open internet databases and the medical terminology understanding of electronic health records (EHRs). We considered that an AI-enhanced system might be also applied to automatically score medical records. This study aimed to develop a series of deep learning models (DLMs) and validated their performance in medical records scoring task. We also analyzed the practical value of the best model. We used the admission medical records from the Tri-Services General Hospital during January 2016 to May 2020, which were scored by our visiting staffs with different levels from different departments. The medical records were scored ranged 0 to 10. All samples were divided into a training set (n = 74,959) and testing set (n = 152,730) based on time, which were used to train and validate the DLMs, respectively. The mean absolute error (MAE) was used to evaluate each DLM performance. In original AI medical record scoring, the predicted score by BERT architecture is closer to the actual reviewer score than the projection word embedding and LSTM architecture. The original MAE is 0.84 ± 0.27 using the BERT model, and the MAE is 1.00 ± 0.32 using the LSTM model. Linear mixed model can be used to improve the model performance, and the adjusted predicted score was closer compared to the original score. However, the project word embedding with the LSTM model (0.66 ± 0.39) provided better performance compared to BERT (0.70 ± 0.33) after linear mixed model enhancement (p &lt, 0.001). In addition to comparing different architectures to score the medical records, this study further uses a mixed linear model to successfully adjust the AI medical record score to make it closer to the actual physician’s score.

Background: glycated hemoglobin (HbA1c) provides information on diabetes mellitus (DM) management. Electrocardiography (ECG) is a noninvasive test of cardiac activity that has been determined to be related to DM and its complications. This study developed a deep learning model (DLM) to estimate HbA1c via ECG. Methods: there were 104,823 ECGs with corresponding HbA1c or fasting glucose which were utilized to train a DLM for calculating ECG-HbA1c. Next, 1539 cases from outpatient departments and health examination centers provided 2190 ECGs for initial validation, and another 3293 cases with their first ECGs were employed to analyze its contributions to DM management. The primary analysis was used to distinguish patients with and without mild to severe DM, and the secondary analysis was to explore the predictive value of ECG-HbA1c for future complications, which included all-cause mortality, new-onset chronic kidney disease (CKD), and new-onset heart failure (HF). Results: we used a gender/age-matching strategy to train a DLM to achieve the best AUCs of 0.8255 with a sensitivity of 71.9% and specificity of 77.7% in a follow-up cohort with correlation of 0.496 and mean absolute errors of 1.230. The stratified analysis shows that DM presented in patients with fewer comorbidities was significantly more likely to be detected by ECG-HbA1c. Patients with higher ECG-HbA1c under the same Lab-HbA1c exhibited worse physical conditions. Of interest, ECG-HbA1c may contribute to the mortality (gender/age adjusted hazard ratio (HR): 1.53, 95% conference interval (CI): 1.08–2.17), new-onset CKD (HR: 1.56, 95% CI: 1.30–1.87), and new-onset HF (HR: 1.51, 95% CI: 1.13–2.01) independently of Lab-HbA1c. An additional impact of ECG-HbA1c on the risk of all-cause mortality (C-index: 0.831 to 0.835, p &lt, 0.05), new-onset CKD (C-index: 0.735 to 0.745, p &lt, 0.01), and new-onset HF (C-index: 0.793 to 0.796, p &lt, 0.05) were observed in full adjustment models. Conclusion: the ECG-HbA1c could be considered as a novel biomarker for screening DM and predicting the progression of DM and its complications.

Context Thyrotoxic periodic paralysis (TPP) characterized by acute weakness, hypokalemia, and hyperthyroidism is a medical emergency with a great challenge in early diagnosis since most TPP patients do not have overt symptoms. Objective This work aims to assess artificial intelligence (AI)-assisted electrocardiography (ECG) combined with routine laboratory data in the early diagnosis of TPP. Methods A deep learning model (DLM) based on ECG12Net, an 82-layer convolutional neural network, was constructed to detect hypokalemia and hyperthyroidism. The development cohort consisted of 39 ECGs from patients with TPP and 502 ECGs of hypokalemic controls; the validation cohort consisted of 11 ECGs of TPP patients and 36 ECGs of non-TPP individuals with weakness. The AI-ECG–based TPP diagnostic process was then consecutively evaluated in 22 male patients with TTP-like features. Results In the validation cohort, the DLM-based ECG system detected all cases of hypokalemia in TPP patients with a mean absolute error of 0.26 mEq/L and diagnosed TPP with an area under curve (AUC) of approximately 80%, surpassing the best standard ECG parameter (AUC = 0.7285 for the QR interval). Combining the AI predictions with the estimated glomerular filtration rate and serum chloride boosted the diagnostic accuracy of the algorithm to AUC 0.986. In the prospective study, the integrated AI and routine laboratory diagnostic system had a PPV of 100% and F-measure of 87.5%. Conclusion An AI-ECG system reliably identifies hypokalemia in patients with paralysis, and integration with routine blood chemistries provides valuable decision support for the early diagnosis of TPP.

The impact of the coronavirus disease 2019 (COVID-19) pandemic on health-care quality in the emergency department (ED) in countries with a low risk is unclear. This study aimed to explore the effects of the COVID-19 pandemic on ED loading, quality of care, and patient prognosis. Data were retrospectively collected from 1 January 2018 to 30 September 2020 at the ED of Tri-service general hospital. Analyses included day-based ED loading, quality of care, and patient prognosis. Data on triage assessment, physiological states, disease history, and results of laboratory tests were collected and analyzed. The number of daily visits significantly decreased after the pandemic, leading to a reduction in the time to examination. Admitted patients benefitted from the pandemic with a reduction of 0.80 h in the length of stay in the ED, faster discharge without death, and reduced re-admission. However, non-admitted visits with chest pain increased the risk of mortality after the pandemic. In conclusion, the COVID-19 pandemic led to a significant reduction in low-acuity ED visits and improved prognoses for hospitalized patients. However, clinicians should be alert about patients with chest pain due to their increased risk of mortality in subsequent admission.

Background: The association between attention-deficit hypersensitivity disorder (ADHD) and the risk of developing colorectal cancer (CRC) is, as yet, to be investigated, and thus, we have conducted this nationwide, cohort study to examine the association in patients from Taiwan.Methods: In this study, 798 individuals with newly diagnosed ADHD and 2,394 (1:3) age-, gender-, and index year- matched controls without ADHD were enrolled, between 2000 and 2013, from the Longitudinal Health Insurance Database, a subset of the National Health Insurance Research Database in Taiwan. The cumulative incidence of CRC was assessed in each cohort by the Kaplan–Meier method. The multivariate Cox proportional hazards model was used to estimate the crude, and the adjusted hazards ratios (HRs) with 95% confidence intervals (CIs), was conducted to estimate the association between ADHD and CRC.Results: The Kaplan–Meier analysis revealed that the cumulative incidence of CRC was significantly higher in patients with ADHD than in those without it (log rank test, p < 0.001). After adjustments for age, gender, comorbidities, and other covariates, the ADHD group was associated with an increased risk of CRC in comparison to the non-ADHD group (adjusted HR = 3.458, 95% CI = 1.640–7.293, p < 0.001). In addition, the usage of methylphenidate was not associated with the risk of developing CRC in patients with ADHD.Conclusion: This retrospective cohort study depicts the evidence that ADHD was associated with the increased risk of CRC. Further studies are needed to confirm the association and the underlying mechanisms.

Federated learning (FL) is a method used for training artificial intelligence models with data from multiple sources while maintaining data anonymity, thus removing many barriers to data sharing. Here we used data from 20 institutes across the globe to train a FL model, called EXAM (electronic medical record (EMR) chest X-ray AI model), that predicts the future oxygen requirements of symptomatic patients with COVID-19 using inputs of vital signs, laboratory data and chest X-rays. EXAM achieved an average area under the curve (AUC)0.92 for predicting outcomes at 24 and 72 h from the time of initial presentation to the emergency room, and it provided 16% improvement in average AUC measured across all participating sites and an average increase in generalizability of 38% when compared with models trained at a single site using that site's data. For prediction of mechanical ventilation treatment or death at 24 h at the largest independent test site, EXAM achieved a sensitivity of 0.950 and specificity of 0.882. In this study, FL facilitated rapid data science collaboration without data exchange and generated a model that generalized across heterogeneous, unharmonized datasets for prediction of clinical outcomes in patients with COVID-19, setting the stage for the broader use of FL in healthcare.

(1) Background: It has rarely been studied whether the severity of diabetic retinopathy (DR) could influence renal disease progression in end-stage renal disease (ESRD) and chronic kidney disease (CKD) in patients with type 2 diabetes. The aim of this study was to evaluate renal disease progression in ESRD and CKD according to DR severity in patients with type 2 diabetes. (2) Methods: We included 1329 patients and divided the cohort into two end-points. The first was to trace the incidence of ESRD in all enrolled participants and the other was to follow their progression to CKD. (3) Results: Significantly higher crude hazard ratios (HRs) of ESRD incidence in all enrolled participants were noted, and this ratio increased in a stepwise fashion. However, after adjustment, DR severity was not associated with ESRD events. Therefore, a subgroup of 841 patients without CKD was enrolled to track their progression to CKD. Compared with no diabetic retinopathy, the progression of CKD increased in a stepwise fashion, from mild nonproliferative diabetic retinopathy (NPDR) to moderate NPDR, to severe NPDR and to proliferative diabetic retinopathy (PDR), both in the crude and adjusted models. (4) Conclusions: The severity of retinopathy appeared to be associated with renal lesions and the development of CKD. Our findings suggest that the severity of DR is a risk factor for progression to CKD. Therefore, diabetic retinopathy is useful for prognosticating the clinical course of diabetic kidney disease.

BackgroundThe initial detection and diagnosis of ST-segment or non-ST-segment elevation myocardial infarction (STEMI or NSTEMI) definitely rely on a 12-lead electrocardiogram (ECG). Delay or misdiagnosis is not unusual by subjective interpretation. Our aim is to develop a DLM as a diagnostic support tool to detect MI based on a 12-lead ECG and to evaluate the performance of this model.MethodsThis study included 1,051 ECGs from 737 coronary angiography (CAG)-validated STEMI patients, 697 ECGs from 287 CAG-validated NSTEMI patients, and 140,336 not-MI ECGs from 76,775 patients at emergency departments. DLM was trained and validated for the performance using 80% and 20% of the ECGs, respectively. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of DLM and experts. STEMI versus not-STEMI, and MI versus not-MI were evaluated by DLM.ResultsThe AUCs of DLM for identifying STEMI and MI were 0.976 and 0.944 in the human-machine competition, respectively, which were significantly better than those of our best clinicians. In the real world setting, DLM presented with AUC of 0.995/0.916 with corresponding sensitivities of 96.9%/77.0%, and specificities of 96.2%/92.9% in the identification of STEMI and MI, respectively. Furthermore, DLM demonstrated sufficient diagnostic capacity for STEMI without the aid of troponin I (TnI) (AUC= 0.996) with corresponding sensitivity and specificity of 98.4% and 96.9%. The AUC of combined DLM and the first recorded TnI for the detection of NSTEMI were increased to 0.978 with corresponding sensitivity and specificity of 91.6% and 96.7%, which was better than that of DLM (0.877) or TnI (0.949) alone. ConclusionsDLM may serve as a diagnostic decision tool to assist intensive or emergency medical system-based networks and frontline physicians in identifying STEMI and NSTEMI in a timely and precise manner to prevent delay or misdiagnosis, and thereby to facilitate subsequent reperfusion therapy.

Programming is one of the most crucial abilities for students in science and technology courses. Few studies on programming ability have considered the effect of students' construal levels on their learning performance. Therefore, the effects of students' construal level were explored in this study to fill this research gap and open a new avenue for the improvements in programming ability. The research participants were 110 seventh- and eighth-grade students with basic programming abilities taking an Arduino course. Data were collected from online questionnaires and analyzed using two-way analysis of variance and structural equation modeling to investigate the relationships among construal levels, programming ability, and learning satisfaction. The results revealed that students' construal levels affect their learning satisfaction and programming ability. These findings indicate that teaching strategies could effectively improve the learning satisfaction and programming ability of junior high school students.

For accurate measurement, phase-shifting technique is usually adopted to the shadow moire measurement system. Accurately introducing the amount of phase shift is required in order to extract the phase properly. However, the specimen or system may be moved during the time of image capture, and not suitable for real-time measurement. In order to overcome this drawback and make an in-line measurement, a shadow moire system consisted of two light source of different colors and a color CCD camera is proposed. The phase shift is introduced by using two light sources illuminate the grating from different position simultaneously. The two moire fringe patterns are captured by the color CCD camera, and are processed by a fringe analysis scheme using spiral phase transform (SPT) and optical flow techniques. The proposed fringe analysis scheme was applied to a simulated surface and a real specimen. The test results are reported and the validity of the scheme is investigated.

The aim of this study was to retrospectively review the clinical outcomes of laparoscopically assisted and open surgical reversal of Hartmann's procedure (HR). We reviewed all patients undergoing laparoscopic or open HR at Tri-Service General Hospital, Taipei, Taiwan, between January 2002 and January 2010. A total of 34 perforated diverticulitis patients initially treated by exploratory laparotomy with Hartmann's procedure were enrolled and divided into 2 groups: laparoscopic and open HR. Data relative to patient age at the time of HR, sex, body mass index, operative time, longest incision length, estimated blood loss, intraoperative complications, postoperative complications, time to bowel function return, duration of hospitalization, and length of follow-up were reviewed. The median colostomy closure period was significantly higher in the laparoscopic group than in the open group (P = 0.011). The median longest incision length, estimated blood loss, time to first oral intake, and hospital stay were significantly lower in the laparoscopic group compared with the open group. Laparoscopic HR may be a technically safe, feasible approach that provides better cosmesis, less blood loss, and faster recovery compared with open HR.

This study was conducted to evaluate the safety and efficacy of single sebacoyl dinalbuphine ester (SDE) injection (150 mg/2 mL) when administered intramuscularly to patients who underwent hemorrhoidectomy for postoperative long-acting analgesia.A total of 221 patients scheduled for hemorrhoidectomy from 6 centers in Taiwan were randomly divided into SDE group and placebo group, and received the treatment, vehicle or SDE, 1 day before the surgery. Visual analogue scale (VAS) was recorded up to 7 to 10 days. Pain intensity using VAS AUC through 48 hours after surgery was calculated as the primary efficacy endpoint.Area under the curve of VAS pain intensity scores (VAS AUC) through 48 hours after hemorrhoidectomy was significantly less in SDE group than those in placebo group (209.93 vs. 253.53). VAS AUC from the end of surgical procedure to day 7 was also significantly different between SDE and placebo group (630.79 vs. 749.94). SDE group consumed significantly less amount of other analgesics, such as PCA ketorolac and oral ketorolac. Median time from the end of surgery to the first use of pain relief medication was also shortened in the placebo group than in the SDE group. Most adverse events were assessed as mild and tolerable in both groups.SDE injection demonstrated an extended analgesia effect, with a statistically significant reduction in pain intensity through 48 hours and 7 days after hemorrhoidectomy.

Although digoxin is important in heart rate control, the utilization of digoxin is declining due to its narrow therapeutic window. Misdiagnosis or delayed diagnosis of digoxin toxicity is common due to the lack of awareness and the time-consuming laboratory work that is involved. Electrocardiography (ECG) may be able to detect potential digoxin toxicity based on characteristic presentations. Our study attempted to develop a deep learning model to detect digoxin toxicity based on ECG manifestations. This study included 61 ECGs from patients with digoxin toxicity and 177,066 ECGs from patients in the emergency room from November 2011 to February 2019. The deep learning algorithm was trained using approximately 80% of ECGs. The other 20% of ECGs were used to validate the performance of the Artificial Intelligence (AI) system and to conduct a human-machine competition. Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of ECG interpretation between humans and our deep learning system. The AUCs of our deep learning system for identifying digoxin toxicity were 0.912 and 0.929 in the validation cohort and the human-machine competition, respectively, which reached 84.6% of sensitivity and 94.6% of specificity. Interestingly, the deep learning system using only lead I (AUC = 0.960) was not worse than using complete 12 leads (0.912). Stratified analysis showed that our deep learning system was more applicable to patients with heart failure (HF) and without atrial fibrillation (AF) than those without HF and with AF. Our ECG-based deep learning system provides a high-accuracy, economical, rapid, and accessible way to detect digoxin toxicity, which can be applied as a promising decision supportive system for diagnosing digoxin toxicity in clinical practice.

Background It is uncertain whether adjuvant chemotherapy (CMT) improves survival in patients with low-risk Stage II colon cancer. We aimed to determine the disease-free survival (DFS) and 5-year overall survival (OS) of low-risk Stage II colon cancer patients treated with adjuvant tegafur/uracil (UFUR). Methods From January 2004 to December 2011, the follow-up status of 278 low-risk Stage II colon cancer patients who underwent surgery in a single medical center was retrospectively analyzed. These patients were divided into three groups based on whether they received adjuvant CMT with UFUR, adjuvant CMT with 5-fluorouracil, or surgery alone. DFS and 5-year OS curves were calculated using Kaplan–Meier survival analysis and Cox proportional hazards regression. Results In the study population, including 278 low-risk Stage II colon cancer patients with a mean age of 68.28 ± 13.01 years, 132 (47.5%) received adjuvant CMT with UFUR, 49 (17.6%) received adjuvant CMT with 5-fluorouracil, and 97 (34.9%) underwent radical surgery alone. At 5 years, the adjusted DFS and OS of low-risk Stage II colon cancer patients were 85.5% and 81.8%, respectively, in the surgery alone group and 97.9% and 96.2%, respectively, in the surgery plus UFUR > 12 months group ( p = 0.004 and p = 0.098, respectively). In multivariate analysis, CMT with UFUR for more than 12 months increased DFS over surgery alone. There was no statistical difference in the 5-year OS. Conclusion Adjuvant CMT treatment of low-risk Stage II colon cancer patients with UFUR for more than 12 months following surgery improves DFS over surgery alone.

Background: Controversy exists over whether aggressive surgical resection of the primary tumor without metastasectomy first or chemotherapy first in stage IV colon cancer with unresectable synchronous liver-only-metastases (CLM) improves patients. Materials and Methods: We retrospectively reviewed the outcome of 156 patients initially diagnosed with unresectable synchronous CLM who were under treatment in our institution from January 2004 to December 2012. Patients with extrahepatic diseases or previous hepatic resection were excluded. All patients with a follow-up of at least 3 months were included. Progression-free survival (PFS) and 5-year overall survival (OS) curves were calculated using the Kaplan–Meier method. Results: Among the 156 patients with CLM, 43 (27.56%) received aggressive surgical resection of the primary tumor without metastasectomy first, 113 (72.43%) received systemic chemotherapy first. At 5 years, the adjusted PFS and OS in stage IV colon cancer with inoperable metastases were 24.2% and 20.4%, respectively, in the surgical resection of the primary tumor first group and 46.0% and 16.9% in the chemotherapy first group ( P = 0.515 and P = 0.742, respectively). In multivariate analysis, there was no statistical difference in the PFS and 5-year OS between the surgical resection of the primary tumor first group and chemotherapy first. Conclusion: Surgical resection of the primary tumor without metastasectomy first in CLM is not associated with improved survival as compared with chemotherapy first. Additional research is necessary to determine which patients may benefit from this intervention.

Carpal tunnel syndrome (CTS) is a common and sometimes challenging condition, which causes pain; paresthesia; tingling of the thumb, index and long fingers; and even thenar weakness in the hands. The effectiveness of corticosteroid injections seems to be transient, and there is a lack of information around long-term effects in previous studies. In recent years, platelet-rich plasma (PRP) has proven to bean alternative as it encourages tissue regeneration. We hereby describe a patient with CTS showing significant improvements in electrophysiological parameters after receiving PRP injections. The results revealed significant improvements in the distal motor and sensory latencies as well as the sensory nerve action potential and compound muscle action potential amplitudes of the both median nerves. In summary, if patients are refractory to conservative treatments-such as splinting, oral medication, and corticosteroid injection-PRP may be used as an alternative before surgical intervention.

This study aims to develop a compensating method to minimize the shrinkage error of the shell mold (SM) in the injection molding (IM) process to obtain uniform optical power in the central optical zone of soft axial symmetric multifocal contact lenses (CL). The Z-shrinkage error along the Z axis or axial axis of the anterior SM corresponding to the anterior surface of a dry contact lens in the IM process can be minimized by optimizing IM process parameters and then by compensating for additional (Add) powers in the central zone of the original lens design. First, the shrinkage error is minimized by optimizing three levels of four IM parameters, including mold temperature, injection velocity, packing pressure, and cooling time in 18 IM simulations based on an orthogonal array L18 (21×34). Then, based on the Z-shrinkage error from IM simulation, three new contact lens designs are obtained by increasing the Add power in the central zone of the original multifocal CL design to compensate for the optical power errors. Results obtained from IM process simulations and the optical simulations show that the new CL design with 0.1 D increasing in Add power has the closest shrinkage profile to the original anterior SM profile with percentage of reduction in absolute Z-shrinkage error of 55% and more uniform power in the central zone than in the other two cases. Moreover, actual experiments of IM of SM for casting soft multifocal CLs have been performed. The final product of wet CLs has been completed for the original design and the new design. Results of the optical performance have verified the improvement of the compensated design of CLs. The feasibility of this compensating method has been proven based on the measurement results of the produced soft multifocal CLs of the new design. Results of this study can be further applied to predict or compensate for the total optical power errors of the soft multifocal CLs.

In a distributed system, programs interact with each other to achieve its functionality. To build a robust system, developers should care about exception handling. This paper propose a dynamic analysis method to explore exception handling strategies of a third-party library without looking code. In result section, we explore an operation which connects to a MySQL server.

Background In most institutions, locally advanced rectal cancer is treated with neoadjuvant chemoradiotherapy followed by surgery 6–8 weeks later, allowing time for tumor response and recovery from chemoradiotherapy-related toxicities. In our hospital, we continuously administer chemotherapy after the completion of chemoradiotherapy, until 2 weeks before surgery for most patients. Methods This was a retrospective study. Patients received a diagnosis of adenocarcinoma of the rectum at our hospital between January 2003 and December 2008 and received neoadjuvant chemoradiotherapy and curative surgery. Chemoradiotherapy consisted of continuous infusion of 225 mg/m 2 5-fluorouracil, 5 days per week. Radiation therapy was delivered at 1.8 Gy per day, 5 days per week for 5–6 weeks (median radiation dose, 50.4 Gy). Chemotherapy was continued until 2 weeks before surgery, and surgery was performed 6–8 weeks after completion of chemoradiotherapy. Results The study included 119 patients (median age, 61 years; range, 24–84 years). Twenty-nine patients (24.4%) had a complete response and 65 (54.6%) had a partial response. Over a median follow-up duration of 52 months, 10 patients experienced local recurrence and 18 had distant metastasis. The 5-year overall and disease-free survival rates were 80.6% and 72.9%, respectively. Grade 3–4 toxicity only occurred in 14 patients (11.8%). Conclusion Continued chemotherapy with 5-fluorouracil after completing neoadjuvant chemoradiotherapy until 2 weeks before surgery for locally advanced rectal cancer results in a good pathological control rate, with low toxicity. Patients who achieved a complete pathological response had a better long-term oncological outcome than those who did not.

Objective The aim of the study was to compare the effects of corticosteroid injection with lidocaine injection in treating tennis elbow. Design It is a prospective, double-blinded, randomized controlled trial. Patients with tennis elbow for more than 1 mo were recruited from a hospital-based rehabilitation outpatient clinic. A total of 70 patients were recruited, and 61 patients completed the study. Patients received an injection of either 10 mg (1 ml) of triamcinolone (corticosteroid group, n = 30) or 1 ml of 1% lidocaine (lidocaine group, n = 31). All of the outcome measures were evaluated before the intervention and at 2 wks and 2 mos after treatment. Results No significant group differences were observed between the corticosteroid and lidocaine groups regarding Patient-Rated Tennis Elbow Evaluation, Disability of the Arm, Shoulder, and Hand, visual analog scale for pain, and grip strength at baseline and at 2 wks and 2 mos after treatment (P > 0.05). However, within-group comparison showed significant improvement after injection with regard to Patient-Rated Tennis Elbow Evaluation, Disability of the Arm, Shoulder, and Hand, visual analog scale for pain, and grip strength in both groups (P > 0.05). Conclusions No differences in the short-term outcomes were found between lidocaine and corticosteroid injection in a small sample of people with tennis elbow with mean duration of 3.8 mos.

The purpose of this study was to estimate the impact of metronomic therapy with oral tegafur-uracil (UFUR) following an intravenous FOLFOX regimen as surgical adjuvant chemotherapy on the overall survival (OS) and disease-free survival (DFS) of stage III colon cancer patients. From the retrospective database of patients who underwent a surgical resection for colorectal cancer at the Tri-Service General Hospital from October 2008 through December 2014, stage III colon carcinomas treated with radical R0 resection were reviewed. One hundred thirty two patients were treated with a FOLFOX regimen (comparison group), and 113 patients were treated with the same regimen followed by additional oral UFUR (UFUR group). The clinical characteristics and mean age of the comparison and UFUR groups were similar. Furthermore, for all study patients, DFS was not significantly different between the two groups. However, 5-year OS rates were 86.8% and 68.5% in the UFUR and comparison groups, respectively (p = 0.0107). Adding UFUR to a FOLFOX regimen was found to significantly improve the OS in patients with stage III colon cancer. UFUR as a maintenance therapy following FOLFOX regimen as an alternative therapeutic option for the treatment of stage III colon cancer patients.

Background: Acute hemorrhagic rectal ulcer (AHRU) can be found in patients with severe comorbid illness, who are bedridden for a long time. Per anal suturing is a quick and feasible treatment. However, recurrent bleeding occurs frequently after suture ligation of a bleeder and can be life-threatening. However, the risk factor for recurrent bleeding is not well known. Our study tries to clarify the risk factor of repeat AHRU in Intensive Care Unit (ICU) patients. Materials and Methods: From January 2004 to December 2009, the medical records of 32 patients, who were admitted to the ICU of the Tri-Service General Hospital, a tertiary referral center in Taiwan, and who underwent per anal suturing of acute hemorrhagic rectal ulcer were retrospectively reviewed. Results: Of the 96 patients who received emergency treatment for acute massive hematochezia, 32 patients were diagnosed with AHRU. Eight (25%) patients had recurrent bleeding following suture ligation of AHRU and underwent a reoperation; no patient had recurrent bleeding after the second operation. The duration from the first hematochezia attack to surgery ( P = 0.04), liver cirrhosis ( P = 0.002), and coagulopathy ( P = 0.01) were the risk factors of recurrent bleeding after suture ligation of a bleeder. Multivariate logistic regression analysis indicated that liver cirrhosis (OR = 37.77, P = 0.014) was an independent risk factor for recurrent bleeding. Conclusion: AHRU could be a major cause of acute massive hematochezia in patients with severe illness. Our data showed that per anal suturing could quickly and effectively control bleeding. We found that liver cirrhosis was an independent risk factor for recurrent bleeding. Therefore, treatment of a liver cirrhosis patient with AHUR should be more aggressive, such as, early detection and proper suture ligation.

Background: Most current state-of-the-art models for searching the International Classification of Diseases, Tenth Revision Clinical Modification (ICD-10-CM) codes use word embedding technology to capture useful semantic properties. However, they are limited by the quality of initial word embeddings. Word embedding trained by electronic health records (EHRs) is considered the best, but the vocabulary diversity is limited by previous medical records. Thus, we require a word embedding model that maintains the vocabulary diversity of open internet databases and the medical terminology understanding of EHRs. Moreover, we need to consider the particularity of the disease classification, wherein discharge notes present only positive disease descriptions. Objective: We aimed to propose a projection word2vec model and a hybrid sampling method. In addition, we aimed to conduct a series of experiments to validate the effectiveness of these methods. Methods: We compared the projection word2vec model and traditional word2vec model using two corpora sources: English Wikipedia and PubMed journal abstracts. We used seven published datasets to measure the medical semantic understanding of the word2vec models and used these embeddings to identify the three–character-level ICD-10-CM diagnostic codes in a set of discharge notes. On the basis of embedding technology improvement, we also tried to apply the hybrid sampling method to improve accuracy. The 94,483 labeled discharge notes from the Tri-Service General Hospital of Taipei, Taiwan, from June 1, 2015, to June 30, 2017, were used. To evaluate the model performance, 24,762 discharge notes from July 1, 2017, to December 31, 2017, from the same hospital were used. Moreover, 74,324 additional discharge notes collected from seven other hospitals were tested. The F-measure, which is the major global measure of effectiveness, was adopted. Results: In medical semantic understanding, the original EHR embeddings and PubMed embeddings exhibited superior performance to the original Wikipedia embeddings. After projection training technology was applied, the projection Wikipedia embeddings exhibited an obvious improvement but did not reach the level of original EHR embeddings or PubMed embeddings. In the subsequent ICD-10-CM coding experiment, the model that used both projection PubMed and Wikipedia embeddings had the highest testing mean F-measure (0.7362 and 0.6693 in Tri-Service General Hospital and the seven other hospitals, respectively). Moreover, the hybrid sampling method was found to improve the model performance (F-measure=0.7371/0.6698). Conclusions: The word embeddings trained using EHR and PubMed could understand medical semantics better, and the proposed projection word2vec model improved the ability of medical semantics extraction in Wikipedia embeddings. Although the improvement from the projection word2vec model in the real ICD-10-CM coding task was not substantial, the models could effectively handle emerging diseases. The proposed hybrid sampling method enables the model to behave like a human expert.

Background: Automated disease code classification using free-text medical information is important for public health surveillance. However, traditional natural language processing (NLP) pipelines are limited, so we propose a method combining word embedding with a convolutional neural network (CNN). Objective: Our objective was to compare the performance of traditional pipelines (NLP plus supervised machine learning models) with that of word embedding combined with a CNN in conducting a classification task identifying International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes in discharge notes. Methods: We used 2 classification methods: (1) extracting from discharge notes some features (terms, n-gram phrases, and SNOMED CT categories) that we used to train a set of supervised machine learning models (support vector machine, random forests, and gradient boosting machine), and (2) building a feature matrix, by a pretrained word embedding model, that we used to train a CNN. We used these methods to identify the chapter-level ICD-10-CM diagnosis codes in a set of discharge notes. We conducted the evaluation using 103,390 discharge notes covering patients hospitalized from June 1, 2015 to January 31, 2017 in the Tri-Service General Hospital in Taipei, Taiwan. We used the receiver operating characteristic curve as an evaluation measure, and calculated the area under the curve (AUC) and F-measure as the global measure of effectiveness. Results: In 5-fold cross-validation tests, our method had a higher testing accuracy (mean AUC 0.9696; mean F-measure 0.9086) than traditional NLP-based approaches (mean AUC range 0.8183-0.9571; mean F-measure range 0.5050-0.8739). A real-world simulation that split the training sample and the testing sample by date verified this result (mean AUC 0.9645; mean F-measure 0.9003 using the proposed method). Further analysis showed that the convolutional layers of the CNN effectively identified a large number of keywords and automatically extracted enough concepts to predict the diagnosis codes. Conclusions: Word embedding combined with a CNN showed outstanding performance compared with traditional methods, needing very little data preprocessing. This shows that future studies will not be limited by incomplete dictionaries. A large amount of unstructured information from free-text medical writing will be extracted by automated approaches in the future, and we believe that the health care field is about to enter the age of big data. [J Med Internet Res 2017;19(11):e380]

The purpose of this study is to build a cost-effective and easy-to-popularize laparoscopic training system based on improving traditional training box. The system has the capability of objective skills assessment and the function of automatic recording of training process and results, as well as 3-dimensional coordinate tracking of instruments. The results of pilot experiment in laparoscopic-assisted grip skill assessment had been shown the system can assess the different grip ability level between the senior surgeons and junior residents. Regarding to the evaluation of training effectiveness, five subjects without laparoscopic surgery experiences were asked to perform grip training for five days to observe their training curves. According to the experimental results, the total time taken for subject 1 to subject 5 are improved by 54.9%, 52.0%, 60.6%, 23.3%, and 63.5% separately.

This paper presents a modified self-tuning fuzzy controller with union rule base for application in the switched reluctance motor (SRM) drive system. Based on the redevelopment of control rule base, this paper aims to simplify the programme hierarchy and computational complexity of the controller by reducing the number of fuzzy sets in the membership function (MF) without losing the system performance and stability. With the error and change of error as the input variables, the output scaling factor (SF) of the proposed controller can be adjusted continuously via a gain updating factor, whose value is derived from the fuzzy logic reasoning. The controller is implemented based on the dSPACE DS1104 control desk. Experimental results show that the proposed controller is in good agreement with theoretical derivation. Comparisons with the conventional PI-like fuzzy controller (PIFC) demonstrate that the effectiveness of the devised control scheme outperforms its conventional counterpart.