Advancements in Barrett's esophagus detection: The role of artificial intelligence and its implications

Artificial intelligence (AI) is making significant strides in revolutionizing the detection of Barrett's esophagus (BE), a precursor to esophageal adenocarcinoma. In the research article by Tsai et al, researchers utilized endoscopic images to train an AI model, challenging the traditional distinction between endoscopic and histological BE. This approach yielded remarkable results, with the AI system achieving an accuracy of 94.37%, sensitivity of 94.29%, and specificity of 94.44%. The study's extensive dataset enhances the AI model's practicality, offering valuable support to endoscopists by minimizing unnecessary biopsies. However, questions about the applicability to different endoscopic systems remain. The study underscores the potential of AI in BE detection while highlighting the need for further research to assess its adaptability to diverse clinical settings.

ArsenicSkinImageBD: A comprehensive image dataset to classify affected and healthy skin of arsenic-affected people

Compared to other popular research domains, dermatology got less attention among machine learning researchers. One of the main concerns for this problem is an inadequate dataset since collecting samples from the human body is very sensitive. In recent years, arsenic has emerged as a significant issue for dermatologists. Arsenic is a highly toxic substance found in the earth's crust whose small amounts can be very injurious to the human body. People who are exposed to arsenic for a long time through water and food can get cancer and skin lesions. With a view to contributing to this aspect, this dataset has been organized with the help of which the researchers can understand the impact of this contamination and design a solution using artificial intelligence. To the best of our knowledge, this is the first standard, easy-to-use, and open dataset of arsenic diseases. The images were collected from four places in Bangladesh, under the Department of Public Health Engineering, Chapainawabganj, where they are working on arsenic contamination. The dataset has 8892 skin images, with half of them showing people with arsenic effects and the other half showing mixed skin images that are not affected by arsenic. This makes the dataset useful for treating people with arsenic-related conditions. Eventually, this dataset can attract the attention of not only the machine learning researchers, but also scientists, doctors, and other professionals in the associated research field.

An alternative for predicting real-time water levels of urban drainage systems

Storm Water Management Model (SWMM) developed by the United States Environmental Protection Agency (EPA) has been widely applied throughout the world for analysis associated with stormwater runoff, combined sewers, and other drainage facilities. To appropriately manage the runoff in urban areas, an integrated system including the simulations of sewer flow, street flow, and regional channel flow, called the 1D/1D SWMM model, was advocated to be performed. Nevertheless, the execution efficiency of this integrated system still needs to be promoted to meet the demand for real-time forecasting of urban floods. The objective of this study is to seek an alternative for predicting water levels both in the sewer system and on the streets within an urban district during rainstorms by utilizing a dynamic neuron network model. To strengthen the physical structure of the artificial intelligence (AI) model and simultaneously make up for the lack of measured data, simulation results of the 1D/1D SWMM model are provided as labels for the training of the proposed model. The novelty of this research is to propose a new methodology to effectively train the AI model for predicting the spatial distributions of depths based on the hydrologic conditions, geomorphologic properties, as well as the network relation of the drainage system. A two-stage training procedure is proposed in this study to consider more possible inundation conditions in both sewer and street (open channel) drainage networks. The research findings show that the proposed methodology is capable of reaching satisfactory accuracy and assisting the numerical-based SWMM model for real-time warning of drainage systems in the urban district.

Modeling and insights into the structural characteristics of endocrine-disrupting chemicals

Endocrine-disrupting chemicals (EDCs) can cause serious harm to human health and the environment; therefore, it is important to rapidly and correctly identify EDCs. Different computational models have been proposed for the prediction of EDCs over the past few decades, but the reported models are not always easily available, and few studies have investigated the structural characteristics of EDCs. In the present study, we have developed a series of artificial intelligence models targeting EDC receptors: the androgen receptor (AR); estrogen receptor (ER); and pregnane X receptor (PXR). The consensus models achieved good predictive results for validation sets with balanced accuracy values of 87.37%, 90.13%, and 79.21% for AR, ER, and PXR binding assays, respectively. Analysis of the physical-chemical properties suggested that several chemical properties were significantly (p < 0.05) different between EDCs and non-EDCs. We also identified structural alerts that can indicate an EDC, which were integrated into the web server SApredictor. These models and structural characteristics can provide useful tools and information in the discrimination and mechanistic understanding of EDCs in drug discovery and environmental risk assessment.

Diagnostic and Management Performance of ChatGPT in Obstetrics and Gynecology

OBJECTIVES

The use of artificial intelligence (AI) in clinical patient management and medical education has been advancing over time. ChatGPT was developed and trained recently, using a large quantity of textual data from the internet. Medical science is expected to be transformed by its use. The present study was conducted to evaluate the diagnostic and management performance of the ChatGPT AI model in obstetrics and gynecology.

DESIGN

A cross-sectional study was conducted.

PARTICIPANTS/MATERIALS, SETTING, METHODS

This study was conducted in Iran in March 2023. Medical histories and examination results of 30 cases were determined in six areas of obstetrics and gynecology. The cases were presented to a gynecologist and ChatGPT for diagnosis and management. Answers from the gynecologist and ChatGPT were compared, and the diagnostic and management performance of ChatGPT were determined.

RESULTS

Ninety percent (27 of 30) of the cases in obstetrics and gynecology were correctly handled by ChatGPT. Its responses were eloquent, informed, and free of a significant number of errors or misinformation. Even when the answers provided by ChatGPT were incorrect, the responses contained a logical explanation about the case as well as information provided in the question stem.

LIMITATIONS

The data used in this study were taken from the electronic book and may reflect bias in the diagnosis of ChatGPT.

CONCLUSIONS

This is the first evaluation of ChatGPT's performance in diagnosis and management in the field of obstetrics and gynecology. It appears that ChatGPT has potential applications in the practice of medicine and is (currently) free and simple to use. However, several ethical considerations and limitations such as bias, validity, copyright infringement, and plagiarism need to be addressed in future studies.

CXCL-8 as a signature of severe Helicobacter pylori infection and a stimulator of stomach region-dependent immune response

Helicobacter pylori infection is involved in development of diverse gastro-pathologies. Our aim is to investigate potential signature of cytokines-chemokine levels (IL-17A, IL-1β, and CXCL-8) in H. pylori-infected patients and their impact on immune response in both corpus and antrum. Multivariate level analysis with machine learning model were carried out using cytokines/chemokine levels of infected Moroccan patients. In addition, Geo dataset was used to run enrichment analysis following CXCL-8 upregulation. Our analysis showed that combination of cytokines-chemokine levels allowed prediction of positive H. pylori density score with <5% of miss-classification error, with fundus CXCL-8 being the most important variable for this discrimination. Furthermore, CXCL-8 dependent expression profile was mainly associated to IL6/JAK/STAT3 signaling in the antrum, interferons alpha and gamma responses in the corpus and commonly induced transcriptional /proliferative activities. To conclude, CXCL-8 level might be a signature of Moroccan H. pylori-infected patients and an inducer of regional-dependent immune response at the gastric level. Larger trials must be carried out to validate the relevance of these results for diverse populations.