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 describe the constructive process of follow-up of colorectal cancer part in the Database from Colorectal Cancer (DACCA) in West China Hospital. MethodThe article was described in words. ResultsThe specific concepts of follow-up of colorectal cancer including end-stage of follow-up, survival status, follow-up strategy, follow-up emphasis, follow-up plan, follow-up record using communication tools, follow-up frequency, annual follow-up times, and single follow-up record of the DACCA in the West China Hospital were defined. Then they were detailed for their definition, label, structure, error correction, and update. ConclusionThrough the detailed description of the details of follow-up of colorectal cancer of DACCA in West China Hospital, it provides the standard and basis for the clinical application of DACCA in the future, and provides reference for other peers who wish to build a colorectal cancer database.
ObjectiveTo unscramble personal data and its tags and structures of Database from Colorectal Cancer (DACCA) in West China Hospital.MethodThe way of words for description was used.ResultsThe definition and setting of 23 items with 18 categories for the personal data from DACCA in West China Hospital was performed. The relevant data label of each item and the structured way needed at the big data application stage were elaborated and the corrective precautions of classification items were described. The three classification items involved privacy attention were described in detailed.ConclusionsBased on description about personal data from DACCA in West China Hospital, it is provided a clinical standard and guide for analyzing of DACCA in future. It also could provide enough experience for construction of colorectal cancer database by staff from same occupation.
ObjectiveTo explain surgical and medical comorbidities and preoperative physical status of colorectal cancer in detail as well as their tags and structures of Database from Colorectal Cancer (DACCA) in West China Hospital.MethodThe article was described in words.ResultsThe definition to the surgical comorbidities with its related content module, the medical comorbidity with its related content modules, and the preoperative physical status and characteristics of the DACCA in West China Hospital were given. The data label corresponding to each item in the database and the structured way needed for the big data application stage in detail were explained. And the error correction notes for all classification items were described.ConclusionsThrough the detailed description of the medical and surgical comorbidities and the preoperative physical status of DACCA in West China Hospital, it provides the standard and basis for the clinical application of DACCA in the future, and provides reference for other peers who wish to build a colorectal cancer database.
ObjectiveTo analyze the characteristics of colorectal cancer surgery in the current version of Database from Colorectal Cancer (DACCA).MethodsThe DACCA version selected for this data analysis was the updated version on April 16th, 2020. The data items included timing of operation, types of operative procedure, radical resection level of operation, patient’s wish of anus-reserving, types of stomy, date of stoma closure, surgical approaches, extended resection, and type of intersphincteric resection (ISR). The data item interval of stoma closure was added, and the selected data items were statistically analyzed.ResultsThe total number of medical records (data rows) that met the criteria was 11 757, including 2 729 valid data on the timing of operation (23.2%), 11 389 valid data on the types of operative procedure (96.9%), 4 255 valid data on the radical resection level of operation (36.2%), 3 803 valid data on patient’s wish of anus-reserving (32.3%), 4 377 valid data on types of stomy (37.2%), 989 valid data on date of stoma closure (8.4%), 4 418 valid data on surgical approaches (37.6%), 3 941 valid data on extended resection (33.5%), and 1 156 valid data on type of ISR (9.8%). In the timing of operation, the most cases were performed immediately after discovery or neoadjuvant completion (915, 33.5%). In types of operative procedure, ultra low anterior resection (ULAR), right hemicolectomy (RHC), and low anterior resection (LAR) were the most, including 1 986 (17.4%), 1 412 (12.4%), and 1 041 (9.1%) lines. Respectively in the colon and rectal cancer surgery, the proportion of RHC (50.0%) and ULAR (26.0%) was the highest, with 172 (26.1%) and 815 (27.9%) extended resection. In ISR surgery the majority was ISR-2 (741, 64.1%). In radical resection level of operation, the number of R0 was the largest with 2 575 (60.5%) lines. In patient’s wish of anus-reserving, positive and rational were the most with 1 811 (47.6%) and 1 440 (37.9%) lines, respectively. And in types of stomy, there were 2 628 lines (60.0%) without stoma and 1 749 cases (40.0%) with stoma, among which the most lines were right lower ileum stoma (612, 35.0%). The minimum value, maximum value, and median value of interval of stoma closure were 0 d, 2 678 d and 112 d. The linear regression prediction of date of stoma closure by year was \begin{document}${\hat {y}} $\end{document}=9.234 3x+22.394 (R2=0.2928, P=0.07). In the surgical approaches, the majority was standard with 3 182 (72.0%) lines.ConclusionsIn the DACCA, rectal cancer surgery is still the majority, and ULAR is the most type. The application of extended resection in both colon and rectal cancer has important significance. The data related to stoma are diversified and need to be further studied.
ObjectiveTo analyze the follow-up data of colorectal cancer in the Database from Colorectal Cancer (DACCA).MethodsThe information in the Dacca database was screened, and the one whose operative date and follow-up date were not blank in the total data was selected. The follow-up data were analyzed, including length of follow-up, survival outcomes, coping styles (doctors’ attitude and reaction for follow-up), follow-up path (whether to choose out-patient, Wechat, QQ tools, phone call, text message, mobile application, face-to-face), the number of follow-up (the number of out-patient follow-up, the number of telephone follow-up, and the number of follow-up within 5 years).ResultsA total of 6 437 data items were analyzed for colorectal cancer adjuvant follow-up. ① The follow-up period of five years (2004–2015) was 56.6% (3 642/6 437), and the follow-up time was 0–201, 67 (26, 97) months. ② The highest data composition ratio of survival outcomes was “Survival” (79.7%, 4 611/5 787), and in the data with five-year follow-up period (2004–2015), the highest data composition ratio of survival outcomes was “Survival” (75.0%, 2 550/3 401), and the survival rate of the five-year follow-up period in 2008 was the highest (91.4%, 235/257). ③ The highest data composition ratio of the coping styles was the doctors’ active follow-up (76.8%, 2 121/2 762). ④ The highest data composition ratio of the follow-up path was out-patient service (90.6%, 4 236/4 676). ⑤ The highest data composition ratio of the number of out-patient follow-up was conducted by the original surgical team (100%, 4 380/4 380), the specific number was 0–130、5 (2, 10) times. The data composition ratio of telephone follow-up was 86.9% (3 808/4 380) and the specific number was 0–68、0 (0, 1) times. The highest frequency of follow-up was in the first year (89.9%, 3 044/3 386) and the specific number was 0–73、5 (3, 9) times.ConclusionBy expounding the characteristics of the colorectal cancer follow-up from colorectal cancer in DACCA, it provides some references for using big data to determine prognosis.
As an interdisciplinary subject of medicine and artificial intelligence, intelligent diagnosis and treatment has received extensive attention in both academia and industry. Traditional Chinese medicine (TCM) is characterized by individual syndrome differentiation as well as personalized treatment with personality analysis, which makes the common law mining technology of big data and artificial intelligence appear distortion in TCM diagnosis and treatment study. This article put forward an intelligent diagnosis model of TCM, as well as its construction method. It could not only obtain personal diagnosis varying individually through active learning, but also integrate multiple machine learning models for training, so as to form a more accurate model of learning TCM. Firstly, we used big data extraction technique from different case sources to form a structured TCM database under a unified view. Then, taken a pediatric common disease pneumonia with dyspnea and cough as an example, the experimental analysis on large-scale data verified that the TCM intelligent diagnosis model based on active learning is more accurate than the pre-existing machine learning methods, which may provide a new effective machine learning model for studying TCM diagnosis and treatment.
ObjectiveBased on recently update Database from Colorectal Cancer (DACCA), we aimed to analyze the characteristics of in-hospital process management from reginal medical center’s colorectal cancer patients.MethodsWe used Version January 29th, 2019 of DACAA being the analyzing source. The items were included date of first out-patient meeting, admitted date, operative date, discharged date, waiting-time, preoperative staying days, postoperative staying days, hospital staying days, and manage protocol, whose characteristics would be analyzed.ResultsWe left 8 913 lines to be analyzed by filtering DACCA. Useful data lines of first out-patient meeting had 3 915, admitted date had 8 144, operative date had 8 049, and discharged date had 7 958. The average of waiting-time were (9.41±0.43) days, and based on timeline trend for line prediction analyzing, which showed R2=0.101 257, P<0.001. The average of preoperative staying days were (5.41±0.04) days, and based on timeline trend for line prediction analyzing, which showed R2=0.023 671, P<0.001. The average of postoperative staying days were (8.99±0.07) days, and based on timeline trend for line prediction analyzing, which showed R2=0.086 177, P<0.001. The average of hospital staying days were (14.43±0.08) days, and based one timeline trend of line prediction analyzing, which showed R2=0.098 44, P<0.001. Analyzable ERAS data were 2 368 lines in DACCA. Total EARS data in 2 368 lines, there were 108 lines (5%) completed and 2 260 lines (95%) incomplete. Pre/post ERAS data in 2 260 lines, there were 150 lines (7%) completed and 2 110 lines (93%) incomplete. Post ERAS data in 2 110 lines, there were 170 lines (8%) completed and 1 940 lines (92%) incomplete.ConclusionsIn recent 20 years, the regional medical center served in-hospital colorectal cancer patients with decreased preoperative staying days, postoperative staying days, and in-hospital staying days from DACCA analyzing, which could prove the service ability had been in improved. Utilization rate of EARS was increased, and also being the main in-hospital process management.
In the context of informatization and digitization, medical big data has become crucial for promoting medical research and technological innovation, posing unprecedented challenges to the construction and operation of big data research supercomputing platforms. This article systematically elaborates on the construction plan of the scientific research supercomputing platform of the West China Biomedical Big Data Center of Sichuan University, as well as the management and service models that support data research. It also compares the scale and operation of existing scientific research supercomputing platforms at home and abroad, providing a reference for the construction and management of medical big data scientific research supercomputing platforms in other institutions.