Objective To analyze the hot spot and future application trend of artificial intelligence technology in the field of intensive care medicine. Methods The CNKI, WanFang Data, VIP and Web of Science core collection databases were electronically searched to collect the related literature about the application of artificial intelligence in the field of critical medicine from January 1, 2013 to December 31, 2022. Bibliometrics was used to visually analyze the author, country, research institution, co-cited literature and key words. Results A total of 986 Chinese articles and 4 016 English articles were included. The number of articles published had increased year by year in the past decade, and the top three countries in English literature were China, the United States and Germany. The predictive model and machine learning were the most frequent key words in Chinese and English literature, respectively. Predicting disease progression, mortality and prognosis were the research focus of artificial intelligence in the field of critical medicine. ConclusionThe application of artificial intelligence in the field of critical medicine is on the rise, and the research hotspots are mainly related to monitoring, predicting disease progression, mortality, disease prognosis and the classification of disease phenotypes or subtypes.
ObjectiveTo apply the multi-modal deep learning model to automatically classify the ultra-widefield fluorescein angiography (UWFA) images of diabetic retinopathy (DR). MethodsA retrospective study. From 2015 to 2020, 798 images of 297 DR patients with 399 eyes who were admitted to Eye Center of Renmin Hospital of Wuhan University and were examined by UWFA were used as the training set and test set of the model. Among them, 119, 171, and 109 eyes had no retinopathy, non-proliferative DR (NPDR), and proliferative DR (PDR), respectively. Localization and assessment of fluorescein leakage and non-perfusion regions in early and late orthotopic images of UWFA in DR-affected eyes by jointly optimizing CycleGAN and a convolutional neural network (CNN) classifier, an image-level supervised deep learning model. The abnormal images with lesions were converted into normal images with lesions removed using the improved CycleGAN, and the difference images containing the lesion areas were obtained; the difference images were classified by the CNN classifier to obtain the prediction results. A five-fold cross-test was used to evaluate the classification accuracy of the model. Quantitative analysis of the marker area displayed by the differential images was performed to observe the correlation between the ischemia index and leakage index and the severity of DR. ResultsThe generated fake normal image basically removed all the lesion areas while retaining the normal vascular structure; the difference images intuitively revealed the distribution of biomarkers; the heat icon showed the leakage area, and the location was basically the same as the lesion area in the original image. The results of the five-fold cross-check showed that the average classification accuracy of the model was 0.983. Further quantitative analysis of the marker area showed that the ischemia index and leakage index were significantly positively correlated with the severity of DR (β=6.088, 10.850; P＜0.001). ConclusionThe constructed multimodal joint optimization model can accurately classify NPDR and PDR and precisely locate potential biomarkers.
With the development of society and the progress of technology, artificial intelligence (AI) and big data technology have penetrated into all walks of life in social production and promoted social production and lifestyle greatly. In the medical field, the applications of AI, such as AI-assisted diagnosis and treatment, robots, medical imaging and so on, have greatly promoted the development and transformation of the entire medical industry. At present, with the support of national policy, market, and technology, we should seize the opportunity of AI development, so as to build the first-mover advantage of AI development. Of course, the development and challenges are coexisted. In the future development process, we should objectively analyze the gap between our country and developed countries, think about the unfavorable factors such as AI chips and data problems, and extend the application and service of AI and big data to all links of medical industry, integrate with clinic fully, so as to better promote the further development of AI medicine treatment in China.
ObjectiveTo evaluate the quality differences in recommendations generated by large language models (LLM) and clinical practitioners for sarcopenia-related questions. MethodsA sarcopenia knowledge base was constructed based on the latest domestic and international research and consensus guidelines. Using the Python environment, a locally deployed and sarcopenia-focused hybrid vertical LLM (referred to as LC) was implemented via LangChain-LLM. Eight fixed questions covering etiology, diagnosis, and prevention were selected, along with eight virtual patient cases. The evaluation team assessed the quality of answers generated by LC and written by clinical practitioners. Quantitative analysis was performed on the precision, recall, and F1 scores (harmonic mean of precision and recall) of treatment recommendations. ResultsThe responses were generally perceived as "possibly written by humans or AI", with a stronger inclination toward being AI-generated, although the accuracy of such judgments was low. Regarding answer quality attributes, LC's responses were superior to those of clinical practitioners in guideline consistency (P<0.01), exhibited similar acceptability (P>0.05), showed better practicality (P<0.05), and had a lower proportion of "1–2 errors" (P<0.05). Quantitative analysis of treatment recommendations indicated that LC and GPT-4.0 outperformed clinical practitioners in recall and F1 scores (P<0.05), with minimal differences between LC and GPT-4.0. ConclusionThe locally deployed sarcopenia-focused hybrid vertical LLM demonstrates high accuracy and applicability in addressing sarcopenia-related issues, outperforming clinical practitioners and exhibiting strong clinical decision-support capabilities.
Artificial intelligence (AI) has been widely used in all walks of life, including healthcare, and has shown great application value in the auxiliary diagnosis of pulmonary nodules in the medical field. In the face of a large amount of lung imaging data, clinicians use AI tools to identify lesions more quickly and accurately, improving work efficiency, but there are still many problems in this field, such as the high false positive rate of recognition, and the difficulty in identifying special types of nodules. Researchers and clinicians are actively developing and using AI tools to promote their continuous evolution and make them better serve human health. This article reviews the clinical application and research progress of AI-assisted diagnosis of pulmonary nodules.
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia. Early diagnosis and effective management are important to reduce atrial fibrillation‐related adverse events. Photoplethysmography (PPG) is often used to assist wearables for continuous electrocardiograph monitoring, which shows its unique value. The development of PPG has provided an innovative solution to AF management. Serial studies of mobile health technology for improving screening and optimized integrated care in atrial fibrillation have explored the application of PPG in screening, diagnosing, early warning, and integrated management in patients with AF. This review summarizes the latest progress of PPG analysis based on artificial intelligence technology and mobile health in AF field in recent years, as well as the limitations of current research and the focus of future research.
Sample size calculation is an important factor to evaluate the reliability of the diagnostic test. In this paper, a case study of the clinical diagnostic test of artificial intelligence for identification of liver contrast-enhanced ultrasound was performed to conduct two-category and multi-categories studies. Based on sensitivity and specificity, the sample size was then estimated in combination with the statistical characteristics of disease incidence, test level and one/two-sided test. Eventually, the sample size was corrected by integrating the factors of the proportion of training/test dataset and the dropout rate of cases in the medical image recognition system. Moreover, the application of Sample Size Calculator, MedCalc, PASS, and other software can accelerate sample size calculation and reduce the amount of labor.
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%.
Objective To systematically evaluate the accuracy of endoscopy-based artificial intelligence (AI)-assisted diagnostic systems in the diagnosis of early-stage esophageal cancer and provide a scientific basis for its diagnostic value. MethodsPubMed, EMbase, The Cochrane Library, Web of Science, Wanfang database, VIP database and CNKI database were searched by computer to search for the relevant literature about endoscopy-based AI-assisted diagnostic systems for the diagnosis of early esophageal cancer from inception to March 2022. The QUADAS-2 was used for quality evaluation of included studies. Meta-analysis of the literature was carried out using Stata 16, Meta-Disc 1.4 and RevMan 5.4 softwares. A bivariate mixed effects regression model was utilized to calculate the combined diagnostic efficacy of the AI-assisted system and meta-regression analysis was conducted to explore the sources of heterogeneity. ResultsA total of 17 articles were included, which consisted of 13 retrospective cohort studies and 4 prospective cohort studies. The results of the quality evaluation using QUADAS-2 showed that all included literature was of high quality. The obtained meta-analysis results revealed that the AI-assisted system in the diagnosis of esophageal cancer presented a combined sensitivity of 0.94 (95%CI 0.91 to 0.96), a specificity of 0.85 (95%CI 0.74 to 0.92), a positive likelihood ratio of 6.28 (95%CI 3.48 to 11.33), a negative likelihood ratio of 0.07 (95%CI 0.05 to 0.11), a diagnostic odds ratio of 89 (95%CI 38 to 208) and an area under the curve of 0.96 (95%CI 0.94 to 0.98). ConclusionThe AI-assisted diagnostic system has a high diagnostic value for early stage esophageal cancer. However, most of the included studies were retrospective. Therefore, further high-quality prospective studies are needed for validation.
ObjectiveTo evaluate the diagnostic value of artificial intelligence (AI)-assisted diagnostic system for pulmonary cancer based on CT images.MethodsDatabases including PubMed, The Cochrane Library, EMbase, CNKI, WanFang Data and Chinese BioMedical Literature Database (CBM) were electronically searched to collect relevant studies on AI-assisted diagnostic system in the diagnosis of pulmonary cancer from 2010 to 2019. The eligible studies were selected according to inclusion and exclusion criteria, and the quality of included studies was assessed and the special information was identified. Then, meta-analysis was performed using RevMan 5.3, Stata 12.0 and SAS 9.4 softwares. The sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio were pooled and the summary receiver operating characteristic (SROC) curve was drawn. Meta-regression analysis was used to explore the sources of heterogeneity.ResultsTotally 18 studies were included with 4 771 patients. Random effect model was used for the analysis due to the heterogeneity among studies. The results of meta-analysis showed that the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnosis odds ratio and area under the SROC curve were 0.87 [95%CI (0.84, 0.90)], 0.89 [95%CI (0.84, 0.92)], 7.70 [95%CI (5.32, 11.15)], 0.14 [95%CI (0.11, 0.19)], 53.54 [95%CI (30.68, 93.42)] and 0.94 [95%CI (0.91, 0.95)], respectively.ConclusionAI-assisted diagnostic system based on CT images has high diagnostic value for pulmonary cancer, and thus it is worthy of clinical application. However, due to the limited quality and quantity of included studies, above results should be validated by more studies.