CholecTriplet2021: A benchmark challenge for surgical action triplet recognition.

Context-aware decision support in the operating room can foster surgical safety and efficiency by leveraging real-time feedback from surgical workflow analysis. Most existing works recognize surgical activities at a coarse-grained level, such as phases, steps or events, leaving out fine-grained interaction details about the surgical activity; yet those are needed for more helpful AI assistance in the operating room. Recognizing surgical actions as triplets of combination delivers more comprehensive details about the activities taking place in surgical videos. This paper presents CholecTriplet2021: an endoscopic vision challenge organized at MICCAI 2021 for the recognition of surgical action triplets in laparoscopic videos. The challenge granted private access to the large-scale CholecT50 dataset, which is annotated with action triplet information. In this paper, we present the challenge setup and the assessment of the state-of-the-art deep learning methods proposed by the participants during the challenge. A total of 4 baseline methods from the challenge organizers and 19 new deep learning algorithms from the competing teams are presented to recognize surgical action triplets directly from surgical videos, achieving mean average precision (mAP) ranging from 4.2% to 38.1%. This study also analyzes the significance of the results obtained by the presented approaches, performs a thorough methodological comparison between them, in-depth result analysis, and proposes a novel ensemble method for enhanced recognition. Our analysis shows that surgical workflow analysis is not yet solved, and also highlights interesting directions for future research on fine-grained surgical activity recognition which is of utmost importance for the development of AI in surgery.

Self-efficacy of emergency management of domestic helpers in pediatric home accidents: A cross-sectional survey in Hong Kong.

Background: Accidental injuries are the leading cause of deaths and disabilities in children globally and most of them occur at home. To save life and prevent sequelae, domestic helpers (DHs) require providing emergency management (i.e., first aid) to children involved in home accidents. However, their self-efficacy in emergency management for children is rarely investigated. Hence, this study aimed to tap that research gap. Methods: This study adopted a cross-sectional descriptive survey design. A convenience sample of 385 DHs was obtained in Hong Kong. DHs' self-efficacy in emergency management for children involved in home accidents was measured using a 12-item well-validated survey instrument "Self-Efficacy of First Aid in Unintentional Injury at Home". The total score ranged from 0 to 48. A higher score indicates greater confidence in emergency management for children involved in home accidents. Results: All the participants were women and most of them were aged between 31 and 35 years (N = 103, 26.8%). The mean score for DHs' self-efficacy in emergency management was 29.0 (SD 10.1). The three items with the lowest self-efficacy were managing bone fractures, performing cardiopulmonary resuscitation, and providing artificial respiration. Bivariate analysis showed that DHs' self-efficacy was significantly related to their educational level, first aid training, caring experience, and working experience. Multiple linear regression indicated that DHs' educational level (ß = 0.136, p = 0.001) and first aid training (ß = 0.532, p < 0.001) were significantly predicting their self-efficacy. Conclusion: DH's self-efficacy of emergency management for children involved in home accidents was low, particularly in those severe situations and complicated first aid procedures.

Congestive Heart Failure Detection Via Short-Time Electrocardiographic Monitoring For Fast Reference Advice In Urgent Medical Conditions.

This study proposed a detection approach for the congestive heart failure (CHF) by short-time electrocardiographic monitoring. Recent literature only reported that RR intervals and Heart Rate Variability (HRV) indicated key hidden information to discriminate CHF groups from healthy controls. However whether it was possible to find certain short-time electrocardiographic monitoring duration for CHF clinical diagnoses, has not been well addressed. In the study, databases of 54 normal subjects and 15 CHF patients from PhysioNet were introduced. Signals were classified into variable assessment lengths. Based on R-R intervals in the assessment length, raw R-R intervals, mean and standard deviation (STD) of R-R intervals, and clinically standard features of shortterm (5-min) Heart Rate Variability (HRV), were comparatively analyzed, while combining with classifiers of Recurrent Neural Network (RNN), Random Forest (RF), and Support Vector Machine (SVM). The Leave-one-out Cross-Validation (LOOCV) was adopted for performance verification, by which the model extracted from certain assessment length was utilized to test measured data of a subject with the same length. Results showed that based on testing databases, a specific 30-minute duration can be achieved by choosing HRV features in full with sensitivity of 88.55% and specificity of 94.81%. It was believed that a short-time electrocardiographic monitoring for the CHF detection could be feasible if standard HRV features together with the classifier of RF or RNN are adopted. It implied that a short-time electrocardiographic monitoring can be applied for fast reference advice of CHF in urgent medical conditions.

Application of stacked convolutional and long short-term memory network for accurate identification of CAD ECG signals.

Coronary artery disease (CAD) is the most common cause of heart disease globally. This is because there is no symptom exhibited in its initial phase until the disease progresses to an advanced stage. The electrocardiogram (ECG) is a widely accessible diagnostic tool to diagnose CAD that captures abnormal activity of the heart. However, it lacks diagnostic sensitivity. One reason is that, it is very challenging to visually interpret the ECG signal due to its very low amplitude. Hence, identification of abnormal ECG morphology by clinicians may be prone to error. Thus, it is essential to develop a software which can provide an automated and objective interpretation of the ECG signal. This paper proposes the implementation of long short-term memory (LSTM) network with convolutional neural network (CNN) to automatically diagnose CAD ECG signals accurately. Our proposed deep learning model is able to detect CAD ECG signals with a diagnostic accuracy of 99.85% with blindfold strategy. The developed prototype model is ready to be tested with an appropriate huge database before the clinical usage.

Financial conflicts of interest and their association with outcome and quality of fibromyalgia drug therapy randomized controlled trials.

AIMS: To evaluate the association of financial conflicts of interest (FCOI) with the characteristics, outcome and reported methodological quality of fibromyalgia drug therapy randomized controlled trials (FM-RCTs). METHODS: A cross-sectional study of original, parallel-group, drug therapy FM-RCTs published between 1997 and 2011 from Medline and Cochrane Central Register of Controlled Trials was conducted. Two reviewers independently assessed each RCT for funding source, authors' FCOI(s), study characteristics, reporting of methodological measures important for internal validity and outcome (positive [statistically significant result favoring experimental drug for the primary outcome] or non-positive). RESULTS: Forty-seven RCTs were eligible with funding source as: 26 (55.3%) industry; eight (17%) non-profit source(s); five (10.6%) mixed; and eight (17%) unspecified. Industry-funded RCTs were more likely to be multicenter and enroll greater number of patients. Reporting of key methodological measures was suboptimal; however, industry and non-profit funded RCTs did not differ in their reporting. Thirty (63.8%) RCTs had ≥ one author who disclosed an FCOI (receipt of research grant [21, 44.7%], industry sponsor employee [20, 42.6%], receipt of consultancy fee/honorarium [16, 34%] and stock ownership [11, 23.4%]). Although industry funding and certain authors' FCOIs (employment and receipt of consultancy fee/honorarium) were univariately associated with positive outcome, such association was not observed after adjusting for study sample size. CONCLUSIONS: The majority of FM-RCTs were industry-sponsored, and had at least one author with an FCOI. Reporting of key methodological measures was suboptimal. After adjusting for study sample size, no association of industry funding or author's FCOI with study outcome was seen.

Duck liver-associated outbreak of Campylobacteriosis among humans, United Kingdom, 2011.

Campylobacter- spp.-related gastroenteritis in diners at a catering college restaurant was associated with consumption of duck liver pâté. Population genetic analysis indicated that isolates from duck samples were typical of isolates from farmed poultry. Campylobacter spp. contamination of duck liver may present a hazard similar to the increasingly recognized contamination of chicken liver.

Reflecting on pain management for patients with osteoarthritis and other rheumatic disorders: there's more to pain management than managing pain.

SUMMARY Medical progress is measured by advances in science and technology. The pace of discovery will surely accelerate. We are increasingly challenged not only to assimilate new information, but also to reconcile our learning with our art. We present the common clinical problem of managing pain in osteoarthritis as a paradigm for this dilemma in contemporary patient care. We do not yet have the understanding and interventions to do this optimally for all with osteoarthritis, leaving us with uncertainties as we struggle to care for these patients. In a world of growing complexity and sophistication we must not overlook the person who is our patient. It is easy to be seduced by electronic and informational advances, to be entranced by machinery, and to forget the unique individuality and needs of each patient. Osler taught that "the practice of medicine is an art, based on science". This doesn't change.

A strategy for identifying circulating placental RNA markers for fetal growth assessment.

OBJECTIVE: To evaluate whether circulating placental mRNAs in maternal plasma could serve as markers for the assessment of fetal growth or intrauterine growth restriction (IUGR). METHODS: From a panel of placental transcripts detectable in maternal plasma identified by microarray previously, we chose growth-related transcripts, namely CSH1, GH2, KISS1, and ADAM12, as potential growth markers. Relationships between the maternal plasma mRNA concentrations with several fetal growth indicators were studied. Maternal plasma mRNA concentrations from IUGR pregnancies with or without pre-eclampsia (PET) were compared with gestational age matched controls cross-sectionally and longitudinally. The four transcripts were quantified by one-step real-time RT-PCR. RESULTS: Maternal plasma GH2 mRNA significantly correlated with birth weight and fetal biometric measurements. Maternal plasma ADAM12 mRNA concentration was significantly higher in IUGR with PET than normal pregnancies in the cross-sectional comparison. No significant difference was observed for all markers between IUGR without PET and controls in both the cross-sectional and longitudinal comparisons. CONCLUSION: This study presents a potential strategy in identifying surrogate markers for the study of fetal growth. Circulating GH2 mRNA in maternal plasma appeared to be associated with fetal growth. The utility of this strategy and the currently assessed markers could be explored in further studies.