Background Fever of unknown origin (FUO) is a group of diseases with heterogeneous complex causes that are misdiagnosed or have delayed diagnoses. Previous studies have focused mainly on the statistical analysis and research of the cases. The treatments are very different for the different categories of FUO. Therefore, how to intelligently diagnose FUO into one category is worth studying. Objective We aimed to fuse all of the medical data together to automatically predict the categories of the causes of FUO among patients using a machine learning method, which could help doctors diagnose FUO more accurately. Methods In this paper, we innovatively and manually built the FUO intelligent diagnosis (FID) model to help clinicians predict the category of the cause and improve the manual diagnostic precision. First, we classified FUO cases into four categories (infections, immune diseases, tumors, and others) according to the large numbers of different causes and treatment methods. Then, we cleaned the basic information data and clinical laboratory results and structured the electronic medical record (EMR) data using the bidirectional encoder representations from transformers (BERT) model. Next, we extracted the features based on the structured sample data and trained the FID model using LightGBM. Results Experiments were based on data from 2299 desensitized cases from Peking Union Medical College Hospital. From the extensive experiments, the precision of the FID model was 81.68% for top 1 classification diagnosis and 96.17% for top 2 classification diagnosis, which were superior to the precision of the comparative method. Conclusions The FID model showed excellent performance in FUO diagnosis and thus would be a potentially useful tool for clinicians to enhance the precision of FUO diagnosis and reduce the rate of misdiagnosis.

BACKGROUND Fever of Unknown Origin (FUO) is a group of diseases with heterogeneous complex causes, which is used to be misdiagnosed or delayed diagnosed. Previous studies mainly focus on the cases statistical analysis and research. The treatment direction is far different for different categories of FUO. So how to diagnose FUO to one category intelligently is worth studying. OBJECTIVE We would like to fuse all the medical data together to predict the cause category of FUO patients by machine learning method automatically, which could help doctors to diagnose FUO more accurately. METHODS In this paper, we innovatively built the FUO Intelligent Diagnosis (FID) model to help clinicians predict the cause category firstly and improve the precision of diagnosis manually. First, we classified FUO cases into four categories (infections, immune diseases, tumors and others) according to the huge different causes and treatment methods. Then, we cleaned the basic information data, clinical examination results and structured the electronic health records (EHRs) data by BERT model. Next, we extracted features based on the structured sample data and trained the FID by LightGBM. RESULTS Experiments were based on 2299 desensitized cases data from Peking Union Medical College Hospital. By extensive experiments, the precision of FID was 81.7% for the top 1 classification diagnosis, and 96.2% for the top 2 classification diagnosis, which was superior to the comparative methods. Besides, we found for the tumors FUO patients, the average age was higher than that of others and there were more female patients in immune diseases FUO. CONCLUSIONS In conclusion, FID showed excellent performance in FUO diagnosis and it would be meaningful for the clinicians to enhance the precision of FUO diagnosis and reduce the misdiagnosis rate.