Search

To systematically quantify the diagnostic accuracy and identify potential covariates affecting the performance of artificial intelligence (AI) in diagnosing orthopedic fractures.PubMed, Embase, Web of Science, and Cochrane Library were systematically searched for studies on AI applications in diagnosing orthopedic fractures from inception to September 29, 2021. Pooled sensitivity and specificity and the area under the receiver operating characteristic curves (AUC) were obtained. This study was registered in the PROSPERO database prior to initiation (CRD 42021254618).Thirty-nine were eligible for quantitative analysis. The overall pooled AUC, sensitivity, and specificity were 0.96 (95% CI 0.94-0.98), 90% (95% CI 87-92%), and 92% (95% CI 90-94%), respectively. In subgroup analyses, multicenter designed studies yielded higher sensitivity (92% vs. 88%) and specificity (94% vs. 91%) than single-center studies. AI demonstrated higher sensitivity with transfer learning (with vs. without: 92% vs. 87%) or data augmentation (with vs. without: 92% vs. 87%), compared to those without. Utilizing plain X-rays as input images for AI achieved results comparable to CT (AUC 0.96 vs. 0.96). Moreover, AI achieved comparable results to humans (AUC 0.97 vs. 0.97) and better results than non-expert human readers (AUC 0.98 vs. 0.96; sensitivity 95% vs. 88%).AI demonstrated high accuracy in diagnosing orthopedic fractures from medical images. Larger-scale studies with higher design quality are needed to validate our findings.• Multicenter study design, application of transfer learning, and data augmentation are closely related to improving the performance of artificial intelligence models in diagnosing orthopedic fractures. • Utilizing plain X-rays as input images for AI to diagnose fractures achieved results comparable to CT (AUC 0.96 vs. 0.96). • AI achieved comparable results to humans (AUC 0.97 vs. 0.97) but was superior to non-expert human readers (AUC 0.98 vs. 0.96, sensitivity 95% vs. 88%) in diagnosing fractures.

Low health literacy (HL) adversely affects medical adherence and health outcomes in patients with chronic diseases. However, the association between HL and enhanced recovery after surgery (ERAS) adherence and postoperative outcomes has not been investigated in patients undergoing colorectal surgery.The data of all patients from a single academic institution who underwent colorectal surgery on an ERAS pathway from January 2019 to July 2020 were prospectively collected. HL levels were assessed using the Brief Health Literacy Screen (BHLS), a proven tool that was used by surgeons after recruitment. According to the HL score, the participants were categorized into low HL (≤9 points) and high HL (10-15 points) groups. The primary outcome was ERAS adherence. Adherence was measured in 22 perioperative elements, and high adherence was defined as adherence to 17 to 22 elements. Secondary outcomes included postoperative complications, hospital length of stay (LOS), hospital charges, mortality, and readmissions.Of the 865 eligible patients, the high HL group consisted of 329 patients (38.0%), and the low HL group contained 536 patients (62.0%). After propensity score matching (1:1), 240 unique pairs of patients with similar characteristics were selected. Patients with high HL levels had a significantly higher rate of high adherence to ERAS standards than those with low HL levels (55% vs 25.8%; adjusted P.001). In terms of adherence to each item, high HL levels were significantly associated with higher adherence to preoperative optimization (90.8% vs 71.7%; adjusted P.001), postoperative gum chewing (59.2% vs 44.6%; adjusted P = .01), early feeding (59.2% vs 31.3%; adjusted P.001), and early mobilization (56.7% vs 30.4%; adjusted P.001). In the overall study population, adjusted logistic regression analyses also showed that high HL levels were associated with a significantly increased rate of high adherence when compared with low HL levels (adjusted odds ratio [OR], 3.57; 95% confidence interval (CI), 2.50-5.09; P.001). In addition, low HL levels were associated with a significantly higher incidence of postoperative complications (32.1% vs 20.8%; P.01), longer hospital LOS (9 [interquartile range {IQR}, 7-11] vs 7 [IQR, 6-9] d; P.001), and higher hospital charges (10,489 [IQR, 8995-11942] vs 8466 [IQR, 7733-9384] dollar; P.001) among propensity-matched patients. However, there were no differences in the mortality and readmission rates between the HL groups.Low HL levels were associated with lower adherence to ERAS elements among propensity-matched patients undergoing colorectal surgery.

BackgroundMore than 70 percent of the world's population is tortured with neck pain more than once in their vast life, of which 50–85% recur within 1–5 years of the initial episode. With medical resources affected by the epidemic, more and more people seek health-related knowledge via YouTube. This article aims to assess the quality and reliability of the medical information shared on YouTube regarding neck pain.MethodsWe searched on YouTube using the keyword “neck pain” to include the top 50 videos by relevance, then divided them into five and seven categories based on their content and source. Each video was quantitatively assessed using the Journal of American Medical Association (JAMA), DISCERN, Global Quality Score (GQS), Neck Pain-Specific Score (NPSS), and video power index (VPI). Spearman correlation analysis was used to evaluate the correlation between JAMA, GQS, DISCERN, NPSS and VPI. A multiple linear regression analysis was applied to identify video features affecting JAMA, GQS, DISCERN, and NPSS.ResultsThe videos had a mean JAMA score of 2.56 (SD = 0.43), DISCERN of 2.55 (SD = 0.44), GQS of 2.86 (SD = 0.72), and NPSS of 2.90 (SD = 2.23). Classification by video upload source, non-physician videos had the greatest share at 38%, and sorted by video content, exercise training comprised 40% of the videos. Significant differences between the uploading sources were observed for VPI (P = 0.012), JAMA (P < 0.001), DISCERN (P < 0.001), GQS (P = 0.001), and NPSS (P = 0.007). Spearman correlation analysis showed that JAMA, DISCERN, GQS, and NPSS significantly correlated with each other (JAMA vs. DISCERN, p < 0.001, JAMA vs. GQS, p < 0.001, JAMA vs. NPSS, p < 0.001, DISCERN vs. GQS, p < 0.001, DISCERN vs. NPSS, p < 0.001, GQS vs. NPSS, p < 0.001). Multiple linear regression analysis suggested that a higher JAMA score, DISCERN, or GQS score were closely related to a higher probability of an academic, physician, non-physician or medical upload source (P < 0.005), and a higher NPSS score was associated with a higher probability of an academic source (P = 0.001) than of an individual upload source.ConclusionsYouTube videos pertaining to neck pain contain low quality, low reliability, and incomplete information. Patients may be put at risk for health complications due to inaccurate, and incomplete information, particularly during the COVID-19 crisis. Academic groups should be committed to high-quality video production and promotion to YouTube users.

Background Heterotopic ossification (HO) is a common complication after cervical disc replacement (CDR). Biomechanical factors including endplate coverage and intervertebral disc height change may be related to HO formation. However, there is a dearth of quantitative analysis for endplate coverage, intervertebral height change and their combined effects on HO. Methods Patients who underwent single-level or two-level CDR with Prestige-LP were retrospectively reviewed. Clinical outcomes were evaluated through Japanese Orthopaedic Association (JOA) score, Neck Disability Index (NDI) score, and visual analogue scale (VAS) score. Radiological data, including the prosthesis-endplate depth ratio, intervertebral height change, posterior heterotopic ossification (PHO) and angular parameters, were collected. Logistic regression analysis was used to identify the potential risk factors. Receiver operating characteristic curves were plotted and the cut-off values of each potential factors were calculated. Results A total of 138 patients with 174 surgical segments were evaluated. Both the prosthesis-endplate depth ratio (P P P P P Conclusions Inadequate endplate coverage and excessive change of intervertebral height are both potential risk factors for the PHO after CDR. Endplate coverage less than 93.8% or intervertebral height change more than 1.8 mm would increase the risk of PHO. The combination of these two factors may exacerbate the non-uniform distribution of stress in the bone-implant interface and promote HO development.

Itching is a leading symptom of eczema or dermatitis and has a great impact on patients’ lives. Previous studies on itching have focused mostly on atopic dermatitis (AD). A cross-sectional multicenter study was conducted among outpatients with eczema from 39 tertiary hospitals in mainland China from July 1 to September 30, 2014. This work elaborates on itching in different types of eczema. Itching was very common (97%, 8499/8758) in outpatients with eczema. The severity of the itch increased with age and disease duration (P P P P P

Background. The epidemiology of atopic dermatitis (AD) in Chinese outpatients is yet to be clarified.Objectives. To investigate population-based prevalence and clinical features of AD in Chinese outpatients.Methods. A multicenter cross-sectional study was conducted in outpatients with eczema or dermatitis from 39 tertiary hospitals in 15 provinces.Results. This study included 682 patients diagnosed with AD, with the mean age of28.8±20.1years and the median course of5.3±6.9years. AD patients had more severe itching (30.4% versus 13.8%,p<0.001) and clinically suspected bacterial infection (21.7% versus 16.1%,p<0.001) than those of other types of dermatitis. Older patients were more susceptible to have a history of flexion dermatitis (p<0.001), bacterial infection (p=0.005), and severe itching (p<0.001). Outpatients with clinically suspected bacterial infection had 3.53-fold increased risk of AD than those without it (p<0.001). The morbidity rate of AD in the (20–25°N) region is 2.86 times higher than that in the (40–45°N) region [OR (95% CI): 0.352 (0.241–0.514),p<0.001].Conclusions. AD is characterized by unique clinical/demographic features. Bacterial infection and latitude region may have an impact on the incidence of AD in China.

The prevalence of atopic dermatitis (AD) has increased substantially. Previous studies have focused mostly on pediatric patients, while epidemiological investigation on adult AD has been very limited. The aim of this study was to determine the prevalence and clinical features of adult AD in outpatients with dermatitis and eczema in China mainland. A multicenter cross-sectional study was conducted among outpatients with eczema or dermatitis from 39 tertiary hospitals of 15 provinces in China from July 1 to September 30, 2014. Of 8758 patients, 407 were adult AD. Compared with adults with other types of dermatitis, the mean age (41.8 ± 14.3 vs 42.04 ± 15.38 years, P 0.05). The highest (8.7%) and lowest prevalence (3.7%) of adult AD were in 25°N to 30°N and 35°N to 40°N latitude region. A substantial part of adult outpatients with eczema or dermatitis is adult AD. Middle age, more body location involvement, more suspected bacterial infection, and severe itching are the main clinical feathers of adult AD. Geographical environment and economic situation work in synergy to adult AD.

Automated whole breast ultrasound (ABUS) has become a popular screening tool in recent years. To reduce the review time and misdetection from ABUS images by physicians, a computer-aided detection (CADe) system for ABUS images based on a multiview method is proposed in this study. A total of 58 pathology-proven lesions from 41 patients were used to evaluate the performance of the system. In the proposed CADe system, the fuzzy c-mean clustering method was applied to detect tumor candidates from these ABUS images. Subsequently, the tumor likelihoods of these candidates could be estimated by a logistic linear regression model based on the intensity, morphology, location, and size features in the transverse, longitudinal, and coronal views. Finally, the multiview tumor likelihoods of the tumor candidates could be obtained from the estimated tumor likelihoods of the three views, and the tumor candidates with high multiview tumor likelihoods were regarded as the detected tumors in the proposed system. The sensitivities of the multiview tumor detection for selecting 5, 10, 20, and 30 tumor candidates with the largest multiview tumor likelihoods were 79.31%, 86.21%, 96.55%, and 98.28%, respectively.

Automated whole breast ultrasound (ABUS) is an emerging screening tool for detecting breast abnormalities. In this study, a computer-aided detection (CADe) system based on multi-scale blob detection was developed for analyzing ABUS images. The performance of the proposed CADe system was tested using a database composed of 136 breast lesions (58 benign lesions and 78 malignant lesions) and 37 normal cases. After speckle noise reduction, Hessian analysis with multi-scale blob detection was applied for the detection of tumors. This method detected every tumor, but some nontumors were also detected. The tumor like lihoods for the remaining candidates were estimated using a logistic regression model based on blobness, internal echo, and morphology features. The tumor candidates with tumor likelihoods higher than a specific threshold (0.4) were considered tumors. By using the combination of blobness, internal echo, and morphology features with 10-fold cross-validation, the proposed CAD system showed sensitivities of 100%, 90%, and 70% with false positives per pass of 17.4, 8.8, and 2.7, respectively. Our results suggest that CADe systems based on multi-scale blob detection can be used to detect breast tumors in ABUS images.

New automated whole breast ultrasound (ABUS) machines have recently been developed and the ultrasound (US) volume dataset of the whole breast can be acquired in a standard manner. The purpose of this study was to develop a novel computer-aided diagnosis system for classification of breast masses in ABUS images. One hundred forty-seven cases (76 benign and 71 malignant breast masses) were obtained by a commercially available ABUS system. Because the distance of neighboring slices in ABUS images is fixed and small, these continuous slices were used for reconstruction as three-dimensional (3-D) US images. The 3-D tumor contour was segmented using the level-set segmentation method. Then, the 3-D features, including the texture, shape and ellipsoid fitting were extracted based on the segmented 3-D tumor contour to classify benign and malignant tumors based on the logistic regression model. The Student's t test, Mann-Whitney U test and receiver operating characteristic (ROC) curve analysis were used for statistical analysis. From the Az values of ROC curves, the shape features (0.9138) are better than the texture features (0.8603) and the ellipsoid fitting features (0.8496) for classification. The difference was significant between shape and ellipsoid fitting features (p = 0.0382). However, combination of ellipsoid fitting features and shape features can achieve a best performance with accuracy of 85.0% (125/147), sensitivity of 84.5% (60/71), specificity of 85.5% (65/76) and the area under the ROC curve Az of 0.9466. The results showed that ABUS images could be used for computer-aided feature extraction and classification of breast tumors.

Purpose: A computer-aided detection (CADe) system based on quantitative tissue clustering algorithm was proposed to identify potential tumors in automated breast ultrasound (ABUS) images. Methods: Our three-dimensional (3D) ABUS images database included 148 biopsy-verified lesions (size 0.4–7.9 cm; mean 1.76 cm). An ABUS volume was comprised of 229–282 slices of two-dimensional (2D) images. For tumor detection, the fast 3D mean shift method was used to remove the speckle noise and the segment tissues with similar properties. The hypoechogenic regions, i.e., the tumor candidates, were extracted using fuzzy c-means clustering. Seven features related to echogenicity and morphology were quantified and used to predict the likelihood of identifying a tumor and filtering out the false-positive (FP) regions. Results: The sensitivity of the proposed CADe system achieved 89.19% (132/148) with 2.00 FPs per volume. For the volumes without lesion, the FP rate was 1.27. The sensitivity was 92.50% (74/80) for malignant tumors and 85.29% (58/68) for benign tumors. Conclusions: The proposed CADe system provides an automatic and quantitative procedure for tumor detection in ABUS images. Further studies are needed to reduce the FP rate of the CADe algorithm.