Image-based molecular representation learning for drug development: a survey.

Artificial intelligence (AI) powered drug development has received remarkable attention in recent years. It addresses the limitations of traditional experimental methods that are costly and time-consuming. While there have been many surveys attempting to summarize related research, they only focus on general AI or specific aspects such as natural language processing and graph neural network. Considering the rapid advance on computer vision, using the molecular image to enable AI appears to be a more intuitive and effective approach since each chemical substance has a unique visual representation. In this paper, we provide the first survey on image-based molecular representation for drug development. The survey proposes a taxonomy based on the learning paradigms in computer vision and reviews a large number of corresponding papers, highlighting the contributions of molecular visual representation in drug development. Besides, we discuss the applications, limitations and future directions in the field. We hope this survey could offer valuable insight into the use of image-based molecular representation learning in the context of drug development.

ENAS-B: Combining ENAS With Bayesian Optimization for Automatic Design of Optimal CNN Architectures for Breast Lesion Classification From Ultrasound Images.

Efficient Neural Architecture Search (ENAS) is a recent development in searching for optimal cell structures for Convolutional Neural Network (CNN) design. It has been successfully used in various applications including ultrasound image classification for breast lesions. However, the existing ENAS approach only optimizes cell structures rather than the whole CNN architecture nor its trainable hyperparameters. This paper presents a novel framework for automatic design of CNN architectures by combining strengths of ENAS and Bayesian Optimization in two-folds. Firstly, we use ENAS to search for optimal normal and reduction cells. Secondly, with the optimal cells and a suitable hyperparameter search space, we adopt Bayesian Optimization to find the optimal depth of the network and optimal configuration of the trainable hyperparameters. To test the validity of the proposed framework, a dataset of 1522 breast lesion ultrasound images is used for the searching and modeling. We then evaluate the robustness of the proposed approach by testing the optimized CNN model on three external datasets consisting of 727 benign and 506 malignant lesion images. We further compare the CNN model with the default ENAS-based CNN model, and then with CNN models based on the state-of-the-art architectures. The results (error rate of no more than 20.6% on internal tests and 17.3% on average of external tests) show that the proposed framework generates robust and light CNN models.

Automatic Detection of Thyroid Nodule Characteristics From 2D Ultrasound Images.

Thyroid cancer is one of the common types of cancer worldwide, and Ultrasound (US) imaging is a modality normally used for thyroid cancer diagnostics. The American College of Radiology Thyroid Imaging Reporting and Data System (ACR TIRADS) has been widely adopted to identify and classify US image characteristics for thyroid nodules. This paper presents novel methods for detecting the characteristic descriptors derived from TIRADS. Our methods return descriptions of the nodule margin irregularity, margin smoothness, calcification as well as shape and echogenicity using conventional computer vision and deep learning techniques. We evaluate our methods using datasets of 471 US images of thyroid nodules acquired from US machines of different makes and labeled by multiple radiologists. The proposed methods achieved overall accuracies of 88.00%, 93.18%, and 89.13% in classifying nodule calcification, margin irregularity, and margin smoothness respectively. Further tests with limited data also show a promising overall accuracy of 90.60% for echogenicity and 100.00% for nodule shape. This study provides an automated annotation of thyroid nodule characteristics from 2D ultrasound images. The experimental results showed promising performance of our methods for thyroid nodule analysis. The automatic detection of correct characteristics not only offers supporting evidence for diagnosis, but also generates patient reports rapidly, thereby decreasing the workload of radiologists and enhancing productivity.

Explainable DCNN Decision Framework for Breast Lesion Classification from Ultrasound Images Based on Cancer Characteristics.

In recent years, deep convolutional neural networks (DCNNs) have shown promising performance in medical image analysis, including breast lesion classification in 2D ultrasound (US) images. Despite the outstanding performance of DCNN solutions, explaining their decisions remains an open investigation. Yet, the explainability of DCNN models has become essential for healthcare systems to accept and trust the models. This paper presents a novel framework for explaining DCNN classification decisions of lesions in ultrasound images using the saliency maps linking the DCNN decisions to known cancer characteristics in the medical domain. The proposed framework consists of three main phases. First, DCNN models for classification in ultrasound images are built. Next, selected methods for visualization are applied to obtain saliency maps on the input images of the DCNN models. In the final phase, the visualization outputs and domain-known cancer characteristics are mapped. The paper then demonstrates the use of the framework for breast lesion classification from ultrasound images. We first follow the transfer learning approach and build two DCNN models. We then analyze the visualization outputs of the trained DCNN models using the EGrad-CAM and Ablation-CAM methods. We map the DCNN model decisions of benign and malignant lesions through the visualization outputs to the characteristics such as echogenicity, calcification, shape, and margin. A retrospective dataset of 1298 US images collected from different hospitals is used to evaluate the effectiveness of the framework. The test results show that these characteristics contribute differently to the benign and malignant lesions' decisions. Our study provides the foundation for other researchers to explain the DCNN classification decisions of other cancer types.

The effect of STAT1, miR-99b, and MAP2K1 in alcoholic liver disease (ALD) mouse model and hepatocyte.

Alcoholic liver disease (ALD) serves as the leading cause of chronic liver diseases-related morbidity and mortality, which threatens the life of millions of patients in the world. However, the molecular mechanisms underlying ALD progression remain unclear. Here, we applied microarray analysis and experimental approaches to identify miRNAs and related regulatory signaling that associated with ALD. Microarray analysis identified that the expression of miR-99b was elevated in the ALD mouse model. The AML-12 cells were treated with EtOH and the expression of miR-99b was enhanced in the cells. The expression of miR-99b was positively correlated with ALT levels in the ALD mice. The microarray analysis identified the abnormally expressed mRNAs in ALD mice and the overlap analysis was performed with based on the differently expressed mRNAs and the transcriptional factors of miR-99b, in which STAT1 was identified. The elevated expression of STAT1 was validated in ALD mice. Meanwhile, the treatment of EtOH induced the expression of STAT1 in the AML-12 cells. The expression of STAT1 was positively correlated with ALT levels in the ALD mice. The positive correlation of STAT1 and miR-99b expression was identified in bioinformatics analysis and ALD mice. The expression of miR-99b and pri-miR-99b was promoted by the overexpression of STAT1 in AML-12 cells. ChIP analysis confirmed the enrichment of STAT1 on miR-99b promoter in AML-12 cells. Next, we found that the expression of mitogen-activated protein kinase kinase 1 (MAP2K1) was negatively associated with miR-99b. The expression of MAP2K1 was downregulated in ALD mice. Consistently, the expression of MAP2K1 was reduced by the treatment of EtOH in AML-12 cells. The expression of MAP2K1 was negative correlated with ALT levels in the ALD mice. We identified the binding site of MAP2K1 and miR-99b. Meanwhile, the treatment of miR-99b mimic repressed the luciferase activity of MAP2K1 in AML-12 cells. The expression of MAP2K1 was suppressed by miR-99b in the cells. We observed that the expression of MAP2K1 was inhibited by the overexpression of STAT1 in AML-12 cells. Meanwhile, the apoptosis of AML-12 cells was induced by the treatment of EtOH, while miR-99b mimic promoted but the overexpression of MAP2K1 attenuated the effect of EtOH in the cells. In conclusion, we identified the correlation and effect of STAT1, miR-99b, and MAP2K1 in ALD mouse model and hepatocyte. STAT1, miR-99b, and MAP2K1 may serve as potential therapeutic target of ALD.

Survival disparities among racial groups with hepatic malignant tumors.

BACKGROUND: Investigating the impact of race on the clinicopathologic characteristics and prognosis of hepatic malignant tumors represents a complex and significant area of research. Notably, distinct differences exist among various racial groups in terms of the clinical manifestations, pathologic features, and prognosis of hepatic malignant tumors. AIM: To explore the effect of race on clinicopathologic features and prognosis of hepatic malignancies. METHODS: Data from patients with hepatic malignancies diagnosed between 2000 and 2019 were collected from the Surveillance, Epidemiology, and End Results database and statistically analyzed. RESULTS: This study included 123558 patients with hepatic malignant tumors, among whom 21078 (17.06%) were Asian, 14810 (11.99%) were Black, and 87670 (70.95%) were white. The median survival times for patients with hepatic malignant tumors of different races were 12.56, 7.70, and 9.35 months for Asian patients, Black patients, and white patients, respectively. The 3-year survival rates for Asian, Black, and white patients were 29%, 19%, and 21%, respectively, and the 5-year survival rates were 22%, 13%, and 15%, respectively. The Kruskal-Wallis test indicated a significant difference in the survival time of patients with hepatic malignant tumors between different races (P < 0.001). Univariate analysis revealed gender disparities in the prognosis among different ethnic groups (Asian: P > 0.05; Black: P < 0.001; White: P < 0.05). Among Black patients, the prognosis was less affected by the degree of hepatic fibrosis than among Asian patients and white patients (Black patients: P < 0.05; Asian patients: P < 0.001; White patients: P < 0.001). Significant differences were observed in the median survival time among patients with hepatic neuroendocrine tumors and hepatoblastomas during pathologic staging between races. Tumor number was inversely related to the prognosis. Cox regression analyses revealed that T stage, M stage, surgery, chemotherapy, alpha-fetoprotein, and tumor size independently influenced prognosis. Age was a specific independent prognostic factor for white patients. Among the tumor stages, N stage is a self-reliant prognostic element specific to white patients. Conversely, radiotherapy and liver fibrosis were not self-reliant prognostic factors for Black patients. Income alone did not independently influence the prognosis of Asian patients. CONCLUSION: The prognosis of hepatic malignant tumors is better among Asian patients than among Black patients. The prognosis of hepatic malignant tumors among white patients is affected by multiple factors, including age and N stage.

Effectiveness of artificial reefs in enhancing phytoplankton community dynamics: A meta-analysis.

Artificial reefs (ARs) are widespread globally and play a positive role in enhancing fish communities and restoring habitat. However, the effect of ARs on phytoplankton, which are fundamental to the marine food chain, remains inconclusive. Conducting a literature review and meta-analysis, this study investigates how ARs influence phytoplankton community dynamics by comparing the biomass, density, and diversity of phytoplankton between ARs and natural water bodies across varying deployment durations, constituent materials, and climatic zones. The study findings suggest that, overall, ARs enhance the biomass, density, and diversity of phytoplankton communities, with no significant differences observed compared to natural water bodies. The enhancement effect of ARs on phytoplankton communities becomes progressively more pronounced with increasing deployment time, with the overall status of phytoplankton communities being optimal when artificial reefs are deployed for 5 years or longer. Concrete and stone ARs can significantly enhance the biomass and diversity of phytoplankton, respectively. The effect of ARs on phytoplankton diversity is unrelated to climatic zones. However, deploying ARs in temperate waters significantly enhances phytoplankton biomass, while in tropical waters, it significantly reduces phytoplankton density. The research findings provide practical implications for the formulation of artificial reef construction strategies tailored to the characteristics of different aquatic ecosystems, emphasizing the need for long-term deployment and appropriate material selection. This study offers a theoretical basis for optimizing AR design and deployment to achieve maximum ecological benefits.

Liver cirrhosis: current status and treatment options using western or traditional Chinese medicine.

Liver cirrhosis arises from liver fibrosis and necroinflammation caused by various mechanisms of hepatic injury. It is a prevalent condition in clinical practice characterized by hepatocellular dysfunction, portal hypertension, and associated complications. Despite its common occurrence, the etiology and pathogenesis of liver cirrhosis remain incompletely understood, posing a significant health threat. Effective prevention of its onset and progression is paramount in medical research. Symptoms often include discomfort in the liver area, while complications such as sarcopenia, hepatic encephalopathy, ascites, upper gastrointestinal bleeding, and infection can arise. While the efficacy of Western medicine in treating liver cirrhosis is uncertain, Chinese medicine offers distinct advantages. This review explores advancements in liver cirrhosis treatment encompassing non-pharmacological and pharmacological modalities. Chinese medicine interventions, including Chinese medicine decoctions, Chinese patent medicines, and acupuncture, exhibit notable efficacy in cirrhosis reversal and offer improved prognoses. Nowadays, the combination of Chinese and Western medicine in the treatment of liver cirrhosis also has considerable advantages, which is worthy of further research and clinical promotion. Standardized treatment protocols based on these findings hold significant clinical implications.

Regression-based Chinese norms of number connection test A and digit symbol test for diagnosing minimal hepatic encephalopathy.

Number connection test A (NCT-A) and digit symbol test (DST), the preferential neuropsychological tests to detect minimal hepatic encephalopathy (MHE) in China, haven't been standardized in Chinese population. We aimed to establish the norms based on a multi-center cross-sectional study and to detect MHE in cirrhotic patients. NCT-A and DST were administered to 648 healthy controls and 1665 cirrhotic patients. The regression-based procedure was applied to develop demographically adjusted norms for NCT-A and DST based on healthy controls. Age, gender, education, and age by education interaction were all predictors of DST, while age, gender, and education by gender interaction were predictors of log10 NCT-A. The predictive equations for expected scores of NCT-A and DST were established, and Z-scores were calculated. The norm for NCT-A was set as Z ≤ 1.64, while the norm for DST was set as Z ≥ - 1.64. Cirrhotic patients with concurrent abnormal NCT-A and DST results were diagnosed with MHE. The prevalence of MHE was 8.89% in cirrhotic patients, and only worse Child-Pugh classification (P = 0.002, OR = 2.389) was demonstrated to be the risk factor for MHE. The regression-based normative data of NCT-A and DST have been developed to detect MHE in China. A significant proportion of Chinese cirrhotic patients suffered from MHE, especially those with worse Child-Pugh classification.

Study on the mechanism of peanut resistance to Fusarium oxysporum infection induced by Bacillus thuringiensis TG5.

Peanut root rot, commonly referred to as rat tail or root rot, is caused by a range of Fusarium species. A strain of bacteria (named TG5) was isolated from crop rhizosphere soil in Mount Taishan, Shandong Province, China, through whole genome sequencing that TG5 was identified as Bacillus thuringiensis, which can specifically produce chloramphenicol, bacitracin, clarithromycin, lichen VK21A1 and bacitracin, with good biological control potential. Based on liquid chromatography tandem mass spectrometry metabonomics analysis and transcriptome conjoint analysis, the mechanism of TG5 and carbendazim inducing peanut plants to resist F. oxysporum stress was studied. In general, for peanut root rot caused by F. oxysporum, B. thuringiensis TG5 has greater advantages than carbendazim and is environmentally friendly. These findings provide new insights for peanut crop genetics and breeding, and for microbial pesticides to replace traditional highly toxic and highly polluting chemical pesticides. Based on the current background of agricultural green cycle and sustainable development, it has significant practical significance and broad application prospects.