Advances in the Application of Traditional Chinese Medicine Using Artificial Intelligence: A Review.

Traditional Chinese medicine (TCM), as one of the crystallizations of Chinese wisdom, emphasizes the balance of Yin and Yang to keep the body healthy. Under the theoretical guidance of a holistic view, the diagnostic process in TCM has characteristics of subjectivity, fuzziness, and complexity. Therefore, realizing standardization and achieving objective quantitative analysis are the bottlenecks of the development of TCM. The emergence of artificial intelligence (AI) technology has brought unprecedented challenges and opportunities to traditional medicine, which is expected to provide objective measurements and improve the clinical efficacy. However, the combination of TCM and AI is still in its infancy and currently faces many challenges. Therefore, this review provides a comprehensive discussion of the existing advances, problems, and prospects of the applications of AI technologies in TCM with the hope of promoting a better understanding of the TCM modernization and intellectualization.

Multi-label classification of fundus images with graph convolutional network and LightGBM.

Early detection and treatment of retinal disorders are critical for avoiding irreversible visual impairment. Given that patients in the clinical setting may have various types of retinal illness, the development of multi-label fundus disease detection models capable of screening for multiple diseases is more in line with clinical needs. This article presented a composite model based on hybrid graph convolution for patient-level multi-label fundus illness identification. The composite model comprised a backbone module, a hybrid graph convolution module, and a classifier module. This article established the relationship between labels via graph convolution and then employed a self-attention mechanism to design a hybrid graph convolution structure. The backbone module extracted features using EfficientNet-B4, whereas the classifier module output multi-label using LightGBM. Additionally, this work investigated the input pattern of binocular images and the influence of label correlation on the model's identification performance. The proposed model MCGL-Net outperformed all other state-of-the-art methods on the publicly available ODIR dataset, with F1 reaching 91.60% on the test set. Ablation experiments were also performed in this paper. Experiments showed that the idea of hybrid graph convolutional structure and composite model designed in this paper promotes the model performance under any backbone CNN. The adoption of hybrid graph convolution can increase the F1 by 2.39% in trials using EfficientNet-B4 as the backbone. The composite model had a higher F1 index by 5.42% than the single EfficientNet-B4 model.

Recognizing lung cancer and stages using a self-developed electronic nose system.

Exhaled breath contains thousands of gaseous volatile organic compounds (VOCs) that could be used as non-invasive biomarkers of lung cancer. Breath-based lung cancer screening has attracted wide attention on account of its convenience, low cost and easy popularization. In this paper, the research of lung cancer detection and staging is conducted by the self-developed electronic nose (e-nose) system. In order to investigate the performance of the device in distinguishing lung cancer patients from healthy controls, two feature extraction methods and two different classification models were adopted. Among all the models, kernel principal component analysis (KPCA) combined with extreme gradient boosting (XGBoost) achieved the best results among 235 breath samples. The accuracy, sensitivity and specificity of e-nose system were 93.59%, 95.60% and 91.09%, respectively. Meanwhile, the device could innovatively classify stages of 90 lung cancer patients (i.e., 44 stage III and 46 stage IV). Experimental results indicated that the recognition accuracy of lung cancer stages was more than 80%. Further experiments of this research also showed that the combination of sensor array and pattern recognition algorithms could identify and distinguish the expiratory characteristics of lung cancer, smoking and other respiratory diseases.

Detection of lung cancer with electronic nose using a novel ensemble learning framework.

Breath analysis based on electronic nose (e-nose) is a promising new technology for the detection of lung cancer that is non-invasive, simple to operate and cost-effective. Lung cancer screening by e-nose relies on predictive models established using machine learning methods. However, using only a single machine learning method to detect lung cancer has some disadvantages, including low detection accuracy and high false negative rate. To address these problems, groups of individual learning models with excellent performance were selected from classic models, including support vector machine, decision tree, random forest, logistic regression andK-nearest neighbor regression, to build an ensemble learning framework (PCA-SVE). The output result of the PCA-SVE framework was obtained by voting. To test this approach, we analyzed 214 breath samples measured by e-nose with 11 gas sensors of four types using the proposed PCA-SVE framework. Experimental results indicated that the accuracy, sensitivity, and specificity of the proposed framework were 95.75%, 94.78%, and 96.96%, respectively. This framework overcomes the disadvantages of a single model, thereby providing an improved, practical alternative for exhaled breath analysis by e-nose.

Lung Cancer Screening Based on Type-different Sensor Arrays.

In recent years, electronic nose (e-nose) systems have become a focus method for diagnosing pulmonary diseases such as lung cancer. However, principles and patterns of sensor responses in traditional e-nose systems are relatively homogeneous. Less study has been focused on type-different sensor arrays. In this paper, we designed a miniature e-nose system using 14 gas sensors of four types and its subsequent analysis of 52 breath samples. To investigate the performance of this system in identifying and distinguishing lung cancer from other respiratory diseases and healthy controls, five feature extraction algorithms and two classifiers were adopted. Lastly, the influence of type-different sensors on the identification ability of e-nose systems was analyzed. Results indicate that when using the LDA fuzzy 5-NN classification method, the sensitivity, specificity and accuracy of discriminating lung cancer patients from healthy controls with e-nose systems are 91.58%, 91.72% and 91.59%, respectively. Our findings also suggest that type-different sensors could significantly increase the diagnostic accuracy of e-nose systems. These results showed e-nose system proposed in this study was potentially practicable in lung cancer screening with a favorable performance. In addition, it is important for type-different sensors to be considered when developing e-nose systems.

Rapid identification and detection of intracellular survival testing of mycobacterium smegmatis mc²155 that contains eis gene from mycobacterium tuberculosis by flow cytometry.

Mycobacterium tuberculosis is a facultative intracellular pathogen that has evolved the ability to survive and multiply within human macrophages. The enhanced intracellular survival (eis) gene (Rv2416c) from M. tuberculosis has been identified as a potential factor that can enhance the intracellular survival of Mycobacterium smegmatis in the macrophage cell line. However, the time requirements for intracellular survival testing of Mycobacterium using classical methodologies are still too long. In this study, we used M. smegmatis mc²155 that contains eis to develop and study a rapid method to test intracellular survival using flow cytometry. We demonstrated the success of this technique, which required only a few hours. This assay is rapid, accurate, and reproducible, and it would be valuable for the rapid detection of intracellular survival of mycobacteria.