A Modified AUC for Training Convolutional Neural Networks: Taking Confidence Into Account

Artificial intelligence applied to magnetic resonance imaging reliably detects the presence, but not the location, of meniscus tears: a systematic review and meta-analysis

OBJECTIVES

To review and compare the accuracy of convolutional neural networks (CNN) for the diagnosis of meniscal tears in the current literature and analyze the decision-making processes utilized by these CNN algorithms.

MATERIALS AND METHODS

PubMed, MEDLINE, EMBASE, and Cochrane databases up to December 2022 were searched in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statement. Risk of analysis was used for all identified articles. Predictive performance values, including sensitivity and specificity, were extracted for quantitative analysis. The meta-analysis was divided between AI prediction models identifying the presence of meniscus tears and the location of meniscus tears.

RESULTS

Eleven articles were included in the final review, with a total of 13,467 patients and 57,551 images. Heterogeneity was statistically significantly large for the sensitivity of the tear identification analysis (I2 = 79%). A higher level of accuracy was observed in identifying the presence of a meniscal tear over locating tears in specific regions of the meniscus (AUC, 0.939 vs 0.905). Pooled sensitivity and specificity were 0.87 (95% confidence interval (CI) 0.80-0.91) and 0.89 (95% CI 0.83-0.93) for meniscus tear identification and 0.88 (95% CI 0.82-0.91) and 0.84 (95% CI 0.81-0.85) for locating the tears.

CONCLUSIONS

AI prediction models achieved favorable performance in the diagnosis, but not location, of meniscus tears. Further studies on the clinical utilities of deep learning should include standardized reporting, external validation, and full reports of the predictive performances of these models, with a view to localizing tears more accurately.

CLINICAL RELEVANCE STATEMENT

Meniscus tears are hard to diagnose in the knee magnetic resonance images. AI prediction models may play an important role in improving the diagnostic accuracy of clinicians and radiologists.

KEY POINTS

• Artificial intelligence (AI) provides great potential in improving the diagnosis of meniscus tears. • The pooled diagnostic performance for artificial intelligence (AI) in identifying meniscus tears was better (sensitivity 87%, specificity 89%) than locating the tears (sensitivity 88%, specificity 84%). • AI is good at confirming the diagnosis of meniscus tears, but future work is required to guide the management of the disease.

Age-Dependent Effects of Homocysteine on Erectile Dysfunction Risk Among U.S. Males: A NHANES Analysis

Erectile dysfunction (ED) is a common problem that seriously impacts men's quality of life and mental health. Earlier studies have indicated that homocysteine (HCY) levels might be linked to the risk of ED, although these studies are limited by small sample sizes and insufficient correction for confounding factors. This study uses data from the 2001-2004 National Health and Nutrition Examination Survey (NHANES) to evaluate the relationship between HCY levels and ED risk in U.S. adult males. The analysis involved using a weighted generalized linear model to assess main effects and restricted cubic splines (RCS) to explore nonlinear relationships. Results showed that the association between HCY and ED was not statistically significant after adjusting for covariates. However, interaction analyses between age and the HCY-ED relationship showed that as age increases, the impact of HCY on ED strengthens. Based on this, subgroup analysis by age was carried out, revealing that in people aged 50 and above, HCY levels were significantly positively correlated with ED, especially when HCY levels exceeded 9.22 μmol/L, significantly increasing the risk of ED. Sensitivity analysis further confirmed the robustness of these findings. This study indicates that controlling HCY levels, especially in middle-aged and older men, might help prevent and treat ED, providing a foundation for future preventive strategies.

Investigating the impact of extreme weather events and related indicators on cardiometabolic multimorbidity

BACKGROUND

The impact of weather on human health has been proven, but the impact of extreme weather events on cardiometabolic multimorbidity (CMM) needs to be urgently explored.

OBJECTIVES

Investigating the impact of extreme temperature, relative humidity (RH), and laboratory testing parameters at admission on adverse events in CMM hospitalizations.

DESIGNS

Time-stratified case-crossover design.

METHODS

A distributional lag nonlinear model with a time-stratified case-crossover design was used to explore the nonlinear lagged association between environmental factors and CMM. Subsequently, unbalanced data were processed by 1:2 propensity score matching (PSM) and conditional logistic regression was employed to analyze the association between laboratory indicators and unplanned readmissions for CMM. Finally, the previously identified environmental factors and relevant laboratory indicators were incorporated into different machine learning models to predict the risk of unplanned readmission for CMM.

RESULTS

There are nonlinear associations and hysteresis effects between temperature, RH and hospital admissions for a variety of CMM. In addition, the risk of admission is higher under low temperature and high RH conditions with the addition of particulate matter (PM, PM2.5 and PM10) and O3_8h. The risk is greater for females and adults aged 65 and older. Compared with first quartile (Q1), the fourth quartile (Q4) had a higher association between serum calcium (HR = 1.3632, 95% CI: 1.0732 ~ 1.7334), serum creatinine (HR = 1.7987, 95% CI: 1.3528 ~ 2.3958), fasting plasma glucose (HR = 1.2579, 95% CI: 1.0839 ~ 1.4770), aspartate aminotransferase/ alanine aminotransferase ratio (HR = 2.3131, 95% CI: 1.9844 ~ 2.6418), alanine aminotransferase (HR = 1.7687, 95% CI: 1.2388 ~ 2.2986), and gamma-glutamyltransferase (HR = 1.4951, 95% CI: 1.2551 ~ 1.7351) were independently and positively associated with unplanned readmission for CMM. However, serum total bilirubin and High-Density Lipoprotein (HDL) showed negative correlations. After incorporating environmental factors and their lagged terms, eXtreme Gradient Boosting (XGBoost) demonstrated a more prominent predictive performance for unplanned readmission of CMM patients, with an average area under the receiver operating characteristic curve (AUC) of 0.767 (95% CI:0.7486 ~ 0.7854).

CONCLUSIONS

Extreme cold or wet weather is linked to worsened adverse health effects in female patients with CMM and in individuals aged 65 years and older. Moreover, meteorologic factors and environmental pollutants may elevate the likelihood of unplanned readmissions for CMM.

Predicting gastric cancer tumor mutational burden from histopathological images using multimodal deep learning

Tumor mutational burden (TMB) is a significant predictive biomarker for selecting patients that may benefit from immune checkpoint inhibitor therapy. Whole exome sequencing is a common method for measuring TMB; however, its clinical application is limited by the high cost and time-consuming wet-laboratory experiments and bioinformatics analysis. To address this challenge, we downloaded multimodal data of 326 gastric cancer patients from The Cancer Genome Atlas, including histopathological images, clinical data and various molecular data. Using these data, we conducted a comprehensive analysis to investigate the relationship between TMB, clinical factors, gene expression and image features extracted from hematoxylin and eosin images. We further explored the feasibility of predicting TMB levels, i.e. high and low TMB, by utilizing a residual network (Resnet)-based deep learning algorithm for histopathological image analysis. Moreover, we developed a multimodal fusion deep learning model that combines histopathological images with omics data to predict TMB levels. We evaluated the performance of our models against various state-of-the-art methods using different TMB thresholds and obtained promising results. Specifically, our histopathological image analysis model achieved an area under curve (AUC) of 0.749. Notably, the multimodal fusion model significantly outperformed the model that relied only on histopathological images, with the highest AUC of 0.971. Our findings suggest that histopathological images could be used with reasonable accuracy to predict TMB levels in gastric cancer patients, while multimodal deep learning could achieve even higher levels of accuracy. This study sheds new light on predicting TMB in gastric cancer patients.

Computer-assisted discovery and evaluation of potential ribosomal protein S6 kinase beta 2 inhibitors

S6K2 is an important protein in mTOR signaling pathway and cancer. To identify potential S6K2 inhibitors for mTOR pathway treatment, a virtual screening of 1,575,957 active molecules was performed using PLANET, AutoDock GPU, and AutoDock Vina, with their classification abilities compared. The MM/PB(GB)SA method was used to identify four compounds with the strongest binding energies. These compounds were further investigated using molecular dynamics (MD) simulations to understand the properties of the S6K2/ligand complex. Due to a lack of available 3D structures of S6K2, OmegaFold served as a reliable 3D predictive model with higher evaluation scores in SAVES v6.0 than AlphaFold, AlphaFold2, and RoseTTAFold2. The 150 ns MD simulation revealed that the S6K2 structure in aqueous solvation experienced compression during conformational relaxation and encountered potential energy traps of about 19.6 kJ mol-1. The virtual screening results indicated that Lys75 and Lys99 in S6K2 are key binding sites in the binding cavity. Additionally, MD simulations revealed that the ligands remained attached to the activation cavity of S6K2. Among the compounds, compound 1 induced restrictive dissociation of S6K2 in the presence of a flexible region, compound 8 achieved strong stability through hydrogen bonding with Lys99, compound 9 caused S6K2 tightening, and the binding of compound 16 was heavily influenced by hydrophobic interactions. This study suggests that these four potential inhibitors with different mechanisms of action could provide potential therapeutic options.

Development of Deep Learning with RDA U-Net Network for Bladder Cancer Segmentation

In today's high-order health examination, imaging examination accounts for a large proportion. Computed tomography (CT), which can detect the whole body, uses X-rays to penetrate the human body to obtain images. Its presentation is a high-resolution black-and-white image composed of gray scales. It is expected to assist doctors in making judgments through deep learning based on the image recognition technology of artificial intelligence. It used CT images to identify the bladder and lesions and then segmented them in the images. The images can achieve high accuracy without using a developer. In this study, the U-Net neural network, commonly used in the medical field, was used to extend the encoder position in combination with the ResBlock in ResNet and the Dense Block in DenseNet, so that the training could maintain the training parameters while reducing the overall identification operation time. The decoder could be used in combination with Attention Gates to suppress the irrelevant areas of the image while paying attention to significant features. Combined with the above algorithm, we proposed a Residual-Dense Attention (RDA) U-Net model, which was used to identify organs and lesions from CT images of abdominal scans. The accuracy (ACC) of using this model for the bladder and its lesions was 96% and 93%, respectively. The values of Intersection over Union (IoU) were 0.9505 and 0.8024, respectively. Average Hausdorff distance (AVGDIST) was as low as 0.02 and 0.12, respectively, and the overall training time was reduced by up to 44% compared with other convolution neural networks.

Predicting colorectal cancer tumor mutational burden from histopathological images and clinical information using multi-modal deep learning

MOTIVATION

Tumor mutational burden (TMB) is an indicator of the efficacy and prognosis of immune checkpoint therapy in colorectal cancer (CRC). In general, patients with higher TMB values are more likely to benefit from immunotherapy. Though whole-exome sequencing is considered the gold standard for determining TMB, it is difficult to be applied in clinical practice due to its high cost. There are also a few DNA panel-based methods to estimate TMB; however, their detection cost is also high, and the associated wet-lab experiments usually take days, which emphasize the need for faster and cheaper alternatives.

RESULTS

In this study, we propose a multi-modal deep learning model based on a residual network (ResNet) and multi-modal compact bilinear pooling to predict TMB status (i.e. TMB high (TMB_H) or TMB low(TMB_L)) directly from histopathological images and clinical data. We applied the model to CRC data from The Cancer Genome Atlas and compared it with four other popular methods, namely, ResNet18, ResNet50, VGG19 and AlexNet. We tested different TMB thresholds, namely, percentiles of 10%, 14.3%, 15%, 16.3%, 20%, 30% and 50%, to differentiate TMB_H and TMB_L.For the percentile of 14.3% (i.e. TMB value 20) and ResNet18, our model achieved an area under the receiver operating characteristic curve of 0.817 after 5-fold cross-validation, which was better than that of other compared models. In addition, we also found that TMB values were significantly associated with the tumor stage and N and M stages. Our study shows that deep learning models can predict TMB status from histopathological images and clinical information only, which is worth clinical application.