Electrocardiogram (ECG) is a noninvasive, inexpensive, and convenient test for diagnosing cardiovascular diseases and assessing the risk of cardiovascular events. Although there are clear standardized operations and procedures for ECG examination, the interpretation of ECG by even trained physicians can be biased due to differences in diagnostic experience. In recent years, artificial intelligence has become a powerful tool to automatically analyze medical data by building deep neural network models, and has been widely used in the field of medical image diagnosis such as CT, MRI, ultrasound and ECG. This article mainly introduces the application progress of deep neural network models in ECG diagnosis and prediction of cardiovascular diseases, and discusses its limitations and application prospects.
In recent years, the computer science represented by artificial intelligence and high-throughput sequencing technology represented by omics play a significant role in the medical field. This paper reviews the research progress of the application of artificial intelligence combined with omics data analysis in the diagnosis and treatment of non-small cell lung cancer (NSCLC), aiming to provide ideas for the development of a more effective artificial intelligence algorithm, and improve the diagnosis rate and prognosis of patients with early NSCLC through a non-invasive way.
ObjectiveTo compare the consistency of artificial analysis and artificial intelligence analysis in the identification of fundus lesions in diabetic patients.MethodsA retrospective study. From May 2018 to May 2019, 1053 consecutive diabetic patients (2106 eyes) of the endocrinology department of the First Affiliated Hospital of Zhengzhou University were included in the study. Among them, 888 patients were males and 165 were females. They were 20-70 years old, with an average age of 53 years old. All patients were performed fundus imaging on diabetic Inspection by useing Japanese Kowa non-mydriatic fundus cameras. The artificial intelligence analysis of Shanggong's ophthalmology cloud network screening platform automatically detected diabetic retinopathy (DR) such as exudation, bleeding, and microaneurysms, and automatically classifies the image detection results according to the DR international staging standard. Manual analysis was performed by two attending physicians and reviewed by the chief physician to ensure the accuracy of manual analysis. When differences appeared between the analysis results of the two analysis methods, the manual analysis results shall be used as the standard. Consistency rate were calculated and compared. Consistency rate = (number of eyes with the same diagnosis result/total number of effective eyes collected) × 100%. Kappa consistency test was performed on the results of manual analysis and artificial intelligence analysis, 0.0≤κ＜0.2 was a very poor degree of consistency, 0.2≤κ＜0.4 meant poor consistency, 0.4≤κ＜0.6 meant medium consistency, and 0.6≤κ＜1.0 meant good consistency.ResultsAmong the 2106 eyes, 64 eyes were excluded that cannot be identified by artificial intelligence due to serious illness, 2042 eyes were finally included in the analysis. The results of artificial analysis and artificial intelligence analysis were completely consistent with 1835 eyes, accounting for 89.86%. There were differences in analysis of 207 eyes, accounting for 10.14%. The main differences between the two are as follows: (1) Artificial intelligence analysis points Bleeding, oozing, and manual analysis of 96 eyes (96/2042, 4.70%); (2) Artificial intelligence analysis of drusen, and manual analysis of 71 eyes (71/2042, 3.48%); (3) Artificial intelligence analyzes normal or vitreous degeneration, while manual analysis of punctate exudation or hemorrhage or microaneurysms in 40 eyes (40/2042, 1.95%). The diagnostic rates for non-DR were 23.2% and 20.2%, respectively. The diagnostic rates for non-DR were 76.8% and 79.8%, respectively. The accuracy of artificial intelligence interpretation is 87.8%. The results of the Kappa consistency test showed that the diagnostic results of manual analysis and artificial intelligence analysis were moderately consistent (κ=0.576, P＜0.01).ConclusionsManual analysis and artificial intelligence analysis showed moderate consistency in the diagnosis of fundus lesions in diabetic patients. The accuracy of artificial intelligence interpretation is 87.8%.
ObjectiveTo summarize the current research progress in the prediction of the efficacy of neoadjuvant therapy of breast cancer based on the application of artificial intelligence (AI) and radiomics. MethodThe researches on the application of AI and radiomics in neoadjuvant therapy of breast cancer in recent 5 years at home and abroad were searched in CNKI, Google Scholar, Wanfang database and PubMed database, and the related research progress was reviewed. ResultsAI had developed rapidly in the field of medical imaging, and molybdenum target, ultrasound and magnetic resonance imaging combined with AI had been deepened and expanded in different degrees in the application research of breast cancer diagnosis and treatment. In the research of molybdenum target combined with AI, the high sensitivity of molybdenum target to microcalcification was mostly used to improve the accuracy of early detection and diagnosis of breast cancer, so as to achieve the clinical purpose of early detection and diagnosis. However, in terms of prediction of neoadjuvant efficacy research of breast cancer, ultrasound and magnetic resonance imaging combined with AI were more prevalent, and their popularity remained unabated. ConclusionIn the monitoring of neoadjuvant therapy for breast cancer, the use of properly designed AI and radiomics models can give full play to its role in the predicting the curative effect of neoadjuvant therapy, and help to guide doctors in clinical diagnosis and treatment and evaluate the prognosis of breast cancer patients.
As an emerging technology, artificial intelligence (AI) uses human theory and technology for robots to study, develop, learn and identify human technologies. Thoracic surgeons should be aware of new opportunities that may affect their daily practice by the direct use of AI technology, or indirect use in the relevant medical fields (radiology, pathology, and respiratory medicine). The purpose of this paper is to review the application status and future development of AI associated with thoracic surgery, diagnosis of AI-related lung cancer, prognosis-assisted decision-making programs and robotic surgery. While AI technology has made rapid progress in many areas, the medical industry only accounts for a small part of AI use, and AI technology is gradually becoming widespread in the diagnosis, treatment, rehabilitation, and care of diseases. The future of AI is bright and full of innovative perspectives. The field of thoracic surgery has conducted valuable exploration and practice on AI, and will receive more and more influence and promotion from AI.
Objective To investigate the accuracy of 18F-FDG positron emission tomography/computed tomography (PET/CT) combined with CT three-dimensional reconstruction (CT-3D) in the differential diagnosis of benign and malignant pulmonary nodules. Methods The clinical data of patients who underwent pulmonary nodule surgery in the Department of Thoracic Surgery, Northern Jiangsu People's Hospital from July 2020 to August 2021 were retrospectively analyzed. The preoperative 18F-FDG PET/CT and chest enhanced CT-3D and other imaging data were extracted. The parameters with diagnostic significance were screened by the area under the receiver operating characteristic (ROC) curve (AUC). Three prediction models, including PET/CT prediction model (MOD PET), CT-3D prediction model (MOD CT-3D), and PET/CT combined CT-3D prediction model (MOD combination), were established through binary logistic regression, and the diagnostic performance of the models were validated by ROC curve. Results A total of 125 patients were enrolled, including 57 males and 68 females, with an average age of 61.16±8.57 years. There were 46 patients with benign nodules, and 79 patients with malignant nodules. A total of 2 PET/CT parameters and 5 CT-3D parameters were extracted. Two PET/CT parameters, SUVmax≥1.5 (AUC=0.688) and abnormal uptake of hilar/mediastinal lymph node metabolism (AUC=0.671), were included in the regression model. Among the CT-3D parameters, CT value histogram peaks (AUC=0.694) and CT-3D morphology (AUC=0.652) were included in the regression model. Finally, the AUC of the MOD PET was verified to be 0.738 [95%CI (0.651, 0.824)], the sensitivity was 74.7%, and the specificity was 60.9%; the AUC of the MOD CT-3D was 0.762 [95%CI (0.677, 0.848)], the sensitivity was 51.9%, and the specificity was 87.0%; the AUC of the MOD combination was 0.857 [95%CI (0.789, 0.925)], the sensitivity was 77.2%, the specificity was 82.6%, and the differences were statistically significant (P<0.001). Conclusion 18F-FDG PET/CT combined with CT-3D can improve the diagnostic performance of pulmonary nodules, and its specificity and sensitivity are better than those of single imaging diagnosis method. The combined prediction model is of great significance for the selection of surgical timing and surgical methods for pulmonary nodules, and provides a theoretical basis for the application of artificial intelligence in the pulmonary nodule diagnosis.
Pathological diagnosis is the gold standard for confirming breast cancer. Traditional manual pathological diagnosis methods for breast cancer are time-consuming, labor-intensive, highly subjective, and exhibit poor diagnostic consistency. In recent years, artificial intelligence (AI) technology has rapidly advanced and is progressively being applied clinically as a promising early diagnostic tool. However, many existing AI models lack interpretability, which limits the trustworthiness of their clinical application. Khater et al, by combining a high-precision machine learning model with an explainable AI model, achieved highly accurate breast tumor diagnosis and provided explanations for key biological and pathological features influencing the diagnostic results. This points the way for the future application and development of AI in medical diagnosis and treatment. The article interprets the main content of that study, and analyzes the advantages and limitations of AI in medical diagnosis and treatment, with the aim of promoting its better application in clinical practice.
ObjectiveTo better understand artificial intelligence (AI) and its application in management of liver cancer.MethodThe relevant literatures about AI in the diagnosis and treatment of liver cancer in recent years were reviewed.ResultsIn terms of diagnosis, the deep learning could precisely and quickly complete the imaging localization and segmentation of the liver, which was helpful for the diagnosis, while radiomics had a high value in assisting the diagnosis of liver cancer, predicting the postoperative recurrence and long-term survival of patients with liver cancer. In regard of treatment, although it was still difficult for AI to intervene in liver surgery, it had significant advantages in formulating individualized operation scheme for patients with liver cancer, which enabled precise hepatectomy and was helpful for prediction of intraoperative bleeding. AI fusion imaging could provide assistance in operation plan making and realize the precise placement of ablation needle. AI was able to predict the tumor response or even tumor progression after interventional therapy and radiotherapy. Pathological analysis was also facilitated by AI and was able to identify some details and feature textures that were difficult to manually distinguish. For transplantation, guidance of AI on the allocation of donor livers based on hazards models helped make better use of limited organ resources. AI could be applied in prognosis prediction in almost all treatment modalities.ConclusionsAI provides more efficient and precise diagnosis, treatment support and prognosis than conventional medical process in liver cancer, generally by constructing a fully functional model based on a series of data mining methods combined with statistical analysis.
ObjectiveTo verify the reliability of Anticlot Assistant, a patient self-management system for warfarin therapy assisted by artificial intelligence.MethodsIt was a single-center, prospective cohort study. The eligible 34 participants were recruited consecutively between November 29, 2017 to September 27, 2018 and managed by warfarin therapy via Anticlot Assistant. The recommendations of Anticlot Assistant were examined and verified by the doctors to ensure the security. Medical records were exported from the the background management system. An univariate analysis compared the outcomes between accepted and overridden records and a logistic regression model was built to determine independent predictors of the outcomes. The research team analyzed 153 medical records, which were from 18 participants and were input by 19 doctors. There were 97 records with doctor accepting the suggestion and 56 records with doctor rejecting the suggestion .ResultsWhen the doctors accepted the recommendations, the percentage of the next-test international normalized ratio (INR) in the therapeutic range was higher (64.95% vs. 44.64%, RR=2.298, 95%CI 1.173 to 4.499, P=0.014). The logistic regression analysis revealed that accepting the recommendations was an independent predictor for the next-test INR being in the therapeutic range after controlling potentially confounding factors (OR=2.446, 95%CI 1.103 to 5.423, P=0.028).ConclusionThe algorithm of Anticlot Assistant is reasonable and reliable.
Artificial intelligence (AI) is reshaping evidence-based clinical decision-making. From the perspective of clinical decision-making, this paper explores the collaborative value of AI in life-cycle health management. While AI can enhance early disease screening efficiency (e.g., medical image analysis) and assist clinical decision-making through personalized health recommendations, its reliance on non-specialized data necessitates the development of dedicated AI systems grounded in high-quality, specialty-specific evidence. AI should serve as an auxiliary tool to evidence-based clinical decision-making, with physicians’ comprehensive judgment and humanistic care remaining central to medical decision-making. Clinicians must improve the reliability of decision making through refining prompt design and cross-validating AI outputs, while actively participate in AI tool optimization and ethical standard development. Future efforts should focus on creating specialty-specific AI tools based on high-quality evidence, establishing dynamic guideline update systems, and formulating medical ethical standards to position AI as a collaborative partner for physicians in implementing life-cycle health management.
