Background: The accuracy of estimating microvascular invasion (MVI) preoperatively in hepatocellular carcinoma (HCC) by clinical observers is low. Most recent studies constructed MVI predictive models utilizing radiological and/or radiomics features extracted from computed tomography (CT) images. These methods, however, rely heavily on human experiences and require manual tumor contouring. We developed a deep learning-based framework for preoperative MVI prediction by using CT images of arterial phase (AP) with simple tumor labeling and without the need of manual feature extraction. The model was further validated on CT images that were originally scanned at multiple different hospitals. Methods: CT images of AP were acquired for 309 patients from China Medical University Hospital (CMUH). Images of 164 patients, who took their CT scanning at 54 different hospitals but were referred to CMUH, were also collected. Deep learning (ResNet-18) and machine learning (support vector machine) models were constructed with AP images and/or patients’ clinical factors (CFs), and their performance was compared systematically. All models were independently evaluated on two patient cohorts: validation set (within CMUH) and external set (other hospitals). Subsequently, explainability of the best model was visualized using gradient-weighted class activation map (Grad-CAM). Results: The ResNet-18 model built with AP images and patients’ clinical factors was superior than other models achieving a highest AUC of 0.845. When evaluating on the external set, the model produced an AUC of 0.777, approaching its performance on the validation set. Model interpretation with Grad-CAM revealed that MVI relevant imaging features on CT images were captured and learned by the ResNet-18 model. Conclusions: This framework provide evidence showing the generalizability and robustness of ResNet-18 in predicting MVI using CT images of AP scanned at multiple different hospitals. Attention heatmaps obtained from model explainability further confirmed that ResNet-18 focused on imaging features on CT overlapping with the conditions used by radiologists to estimate MVI clinically.

Mutant p53 (mutp53) commonly loses its DNA binding affinity to p53 response elements (p53REs) and fails to induce apoptosis fully. However, the p53 mutation does not predict chemoresistance in all subtypes of breast cancers, and the critical determinants remain to be identified. In this study, mutp53 was found to mediate chemotherapy-induced long intergenic noncoding RNA-p21 (lincRNA-p21) expression by targeting the G-quadruplex structure rather than the p53RE on its promoter to promote chemosensitivity. However, estrogen receptor alpha (ERα) suppressed mutp53-mediated lincRNA-p21 expression by hijacking mutp53 to upregulate damaged DNA binding protein 2 (DDB2) transcription for subsequent DNA repair and chemoresistance. Levels of lincRNA-p21 positively correlated with the clinical responses of breast cancer patients to neoadjuvant chemotherapy and had an inverse correlation with the ER status and DDB2 level. In contrast, the carboplatin-induced DDB2 expression was higher in ER-positive breast tumor tissues. These results demonstrated that ER status determines the oncogenic function of mutp53 in chemoresistance by switching its target gene preference from lincRNA-p21 to DDB2 and suggest that induction of lincRNA-p21 and targeting DDB2 would be effective strategies to increase the chemosensitivity of mutp53 breast cancer patients.

Animal coronaviruses (CoVs) have been identified to be the origin of Severe Acute Respiratory Syndrome (SARS)-CoV, Middle East respiratory syndrome (MERS)-CoV, and probably SARS-CoV-2 that cause severe to fatal diseases in humans. Variations of zoonotic coronaviruses pose potential threats to global human beings. To overcome this problem, we focused on the main protease (M^pro), which is an evolutionary conserved viral protein among different coronaviruses. The broad-spectrum anti-coronaviral drug, GC376, was repurposed to target canine coronavirus (CCoV), which causes gastrointestinal infections in dogs. We found that GC376 can efficiently block the protease activity of CCoV M^pro and can thermodynamically stabilize its folding. The structure of CCoV M^pro in complex with GC376 was subsequently determined at 2.75 Å. GC376 reacts with the catalytic residue C144 of CCoV M^pro and forms an (R)- or (S)-configuration of hemithioacetal. A structural comparison of CCoV M^pro and other animal CoV M^pros with SARS-CoV-2 M^pro revealed three important structural determinants in a substrate-binding pocket that dictate entry and release of substrates. As compared with the conserved A141 of the S1 site and P188 of the S4 site in animal coronaviral M^pros, SARS-CoV-2 M^pro contains N142 and Q189 at equivalent positions which are considered to be more catalytically compatible. Furthermore, the conserved loop with residues 46–49 in animal coronaviral M^pros has been replaced by a stable α-helix in SARS-CoV-2 M^pro. In addition, the species-specific dimerization interface also influences the catalytic efficiency of CoV M^pros. Conclusively, the structural information of this study provides mechanistic insights into the ligand binding and dimerization of CoV M^pros among different species.

Since the emergence of COVID-19, many hospitals have encountered challenges in performing efficient scheduling and good resource management to ensure the quality of healthcare provided to patients is not compromised. Operating room (OR) scheduling is one of the issues that has gained our attention because it is related to workflow efficiency and critical care of hospitals. Automatic scheduling and high predictive accuracy of surgical case duration have a critical role in improving OR utilization. To estimate surgical case duration, many hospitals rely on historic averages based on a specific surgeon or a specific procedure type obtained from electronic medical record (EMR) scheduling systems. However, the low predictive accuracy with EMR data leads to negative impacts on patients and hospitals, such as rescheduling of surgeries and cancellation. In this study, we aim to improve the prediction of surgical case duration with advanced machine learning (ML) algorithms. We obtained a large data set containing 170,748 surgical cases (from Jan 2017 to Dec 2019) from a hospital. The data covered a broad variety of details on patients, surgeries, specialties and surgical teams. In addition, a more recent data set with 8,672 cases (from Mar to Apr 2020) was available to be used for external evaluation. We computed historic averages from the EMR data for surgeon- or procedure-specific cases, and they were used as baseline models for comparison. Subsequently, we developed our models using linear regression, random forest and extreme gradient boosting (XGB) algorithms. All models were evaluated with R-square (R²), mean absolute error (MAE), and percentage overage (actual duration longer than prediction), underage (shorter than prediction) and within (within prediction). The XGB model was superior to the other models, achieving a higher R² (85 %) and percentage within (48 %) as well as a lower MAE (30.2 min). The total prediction errors computed for all models showed that the XGB model had the lowest inaccurate percentage (23.7 %). Overall, this study applied ML techniques in the field of OR scheduling to reduce the medical and financial burden for healthcare management. The results revealed the importance of surgery and surgeon factors in surgical case duration prediction. This study also demonstrated the importance of performing an external evaluation to better validate the performance of ML models.

Predictive accuracy of surgical case duration plays a critical role in reducing cost of operation room (OR) utilization. The most common approaches used by hospitals rely on historic averages based on a specific surgeon or a specific procedure type obtained from the electronic medical record (EMR) scheduling systems. However, low predictive accuracy of EMR leads to negative impacts on patients and hospitals, such as rescheduling of surgeries and cancellation. In this study, we aim to improve prediction of operation case duration with advanced machine learning (ML) algorithms. We obtained a large data set containing 170,748 operation cases (from Jan 2017 to Dec 2019) from a hospital. The data covered a broad variety of details on patients, operations, specialties and surgical teams. Meanwhile, a more recent data with 8,672 cases (from Mar to Apr 2020) was also available to be used for external evaluation. We computed historic averages from EMR for surgeon- or procedure-specific and they were used as baseline models for comparison. Subsequently, we developed our models using linear regression, random forest and extreme gradient boosting (XGB) algorithms. All models were evaluated with R-squre ( R ² ), mean absolute error ( MAE ), and percentage overage (case duration > prediction + 10 % & 15 mins), underage (case duration < prediction - 10 % & 15 mins) and within (otherwise). The XGB model was superior to the other models by having higher R ² (85 %) and percentage within (48 %) as well as lower MAE (30.2 mins). The total prediction errors computed for all the models showed that the XGB model had the lowest inaccurate percent (23.7 %). As a whole, this study applied ML techniques in the field of OR scheduling to reduce medical and financial burden for healthcare management. It revealed the importance of operation and surgeon factors in operation case duration prediction. This study also demonstrated the importance of performing an external evaluation to better validate performance of ML models.