When the epidemiology of infectious diseases is more complex, it is often difficult for disease prediction studies based on a single model to capture the multidimensional nature of disease transmission. In recent years, combining different models to improve infectious disease prediction has gradually become a research trend and hotspot. Existing studies have shown that combined models usually have higher prediction performance and better generalization ability. The current combined models mainly combine machine learning and other models, including time-series models, dynamic models, etcetera. In addition, integrated learning that combines diverse machine learning techniques also holds significant importance across various research domains. This paper reviews the progress of applying combined models around machine learning in infectious disease prediction to promote the innovation and practice of combined models for infectious diseases and help to build smarter and more efficient infectious disease early warning and prediction methods and systems.

Methods such as compartmental models, agent-based models, time series models, and machine learning can be used for the prediction of infectious disease incidence. When disease epidemics are complex, it is often difficult to use a single model to comprehensively and accurately capture the multi dimensional nature of the disease. Exploring the combined application of different models has gradually become a research trend and hotspot in recent years, and the prediction performance of combined models is often better than that of single ones. Current research related to combined models mainly focus on machine learning or compartmental models. In this review, we focus on the combination of compartmental models and other models, and summarize their combination principles, application progress, and advantages or disadvantages for the purpose of promoting the innovation and application of combined models for infectious disease incidence prediction, and establishing a more intelligent and efficient early warning and prediction method or systems for the prevention and control of infectious disease.

Methods such as compartmental models, agent-based models, time series models, and machine learning can be used for the prediction of infectious disease incidence. When disease epidemics are complex, it is often difficult to use a single model to comprehensively and accurately capture the multi dimensional nature of the disease. Exploring the combined application of different models has gradually become a research trend and hotspot in recent years, and the prediction performance of combined models is often better than that of single ones. Current research related to combined models mainly focus on machine learning or compartmental models. In this review, we focus on the combination of compartmental models and other models, and summarize their combination principles, application progress, and advantages or disadvantages for the purpose of promoting the innovation and application of combined models for infectious disease incidence prediction, and establishing a more intelligent and efficient early warning and prediction method or systems for the prevention and control of infectious disease.

Objective:To analyze the clinical characteristics, RAN-binding protein 2 ( RANBP2) gene variations, and prognosis in Chinese acute necrotizing encephalopathy (ANE) patients. Methods:A retrospective analysis of ANE cases registered in the Peking Union Medical College Hospital Encephalitis Registry System from 2022 to 2024, involving patients from Peking Union Medical College Hospital and other hospitals, was conducted. A descriptive study was performed on the clinical characteristics, treatments and prognosis, cerebrospinal fluid examination results, and imaging findings of these patients based on adjusted ANE diagnostic criteria. Whole-exome sequencing technology was used to detect gene mutations in these patients.Results:A total of 18 ANE cases were included, ranged in age from 2 to 72 [20(5, 43)] years. The male-to-female ratio was 4∶5. All patients were found with precipitating infections including COVID-19, influenza A virus and Mycoplasma pneumoniae infections. All patients presented with fever, with varying degrees of consciousness disturbance observed in 16 cases, and seizures in 10 cases. All patients underwent lumbar puncture, with normal or mildly elevated white cell counts [3(2, 13)×10 6/L] and mildly to moderately elevated protein levels [1.90(0.92, 4.65) g/L]. A total of 6 patients were found with extremely elevated interleukin-6 level [950(164, 2 000) pg/ml] in cerebrospinal fluid. Bilateral symmetric thalamic lesions were typical imaging features of ANE, while involvement of other areas such as cortical and subcortical white matter, brainstem, and cerebellum was also observed. A total of 14 patients performed genetic tests while 4 patients were identified with RANBP2 gene mutations (c.1754C>T in 3 cases, c.1966A>G in 1 case). All patients received immunotherapy, and 7 patients died at discharge while other patients presented with neurological sequelae of varying degrees. Conclusions:ANE is a rare and severe parainfectious encephalopathy that can occur in both children and adults. Clinically, it is characterized by rapidly progressing encephalopathy following systematic infection, with bilateral symmetric thalamic lesions. The detection of RANBP2 gene mutations could help make the diagnosis.

Objective:To analyze the efficacy of laparoscopic partial splenectomy and devascularization (LPSPD), open partial splenectomy and devascularization (OPSPD), laparoscopic total splenectomy and devascularization (LTSPD), and open total splenectomy and devascularization (OTSPD) in the treatment of portal hypertension.Methods:Clinical data of 128 patients with portal hypertension admitted to the Department of Hepatobiliary and Pancreatic Surgery of the Affiliated Hospital of Qinghai University and the General Surgery Department of the Fifth Medical Center of the Chinese PLA General Hospital from May 2018 to May 2022 were retrospectively analyzed, including 69 males and 59 females, aged 51.0±10.9 years. According to different surgical methods, patients were divided into four groups, namely the LPSPD group ( n=30), the OPSPD group ( n=32), the LTSPD group ( n=31), and the OTSPD group ( n=35). Operation time, postoperative hospitalization time, postoperative complications such as newly occurred PVT, etc of the four groups of patients were compared. Results:The operation times of the LPSPD, OPSPD, LTSPD and OTSPD group were 240.0 (180.0, 399.0), 209.5 (169.0, 259.0), 207.0 (175.3, 297.5) and 200.0 (162.0, 264.0) minutes, respectively. The postoperative hospital stays were 8.0 (7.0, 12.0), 10.0 (8.3, 11.8), 11.0 (9.0, 13.0) and 13.0 (11.0, 15.0) days, respectively. The incidences of newly occurred PVT after surgery were 3.3% (1/30), 3.1% (1/32), 19.4% (6/31), and 20.0% (7/35), respectively. The above indicators were compared among the four groups respectively, and the differences were statistically significant (all P<0.05). The postoperative hospital stay of the LPSPD group was shorter than that of the LTSPD group [8.0 (7.0, 12.0) vs 11.0 (9.0, 13.0), d, Z=-3.93], and the postoperative hospital stay of the OPSPD group was also shorter than that of the OTSPD group [10.0 (8.3, 11.8) vs 13.0 (11.0, 15.0), d, Z=-3.56; all P<0.001]. The incidence of newly occurred PVT after surgery in the partial splenectomy group (LPSPD and OPSPD) was 3.2% (2/62), which was lower than that in the total splenectomy group [LTSPD and OTSPD, 19.7% (13/66), χ2=8.38, P=0.004]. Conclusion:Compared to total splenectomy, partial splenectomy might be favorable for the shortened postoperative hospital stay and lower incidence of newly occurred portal vein thrombosis.

When the epidemiology of infectious diseases is more complex, it is often difficult for disease prediction studies based on a single model to capture the multidimensional nature of disease transmission. In recent years, combining different models to improve infectious disease prediction has gradually become a research trend and hotspot. Existing studies have shown that combined models usually have higher prediction performance and better generalization ability. The current combined models mainly combine machine learning and other models, including time-series models, dynamic models, etcetera. In addition, integrated learning that combines diverse machine learning techniques also holds significant importance across various research domains. This paper reviews the progress of applying combined models around machine learning in infectious disease prediction to promote the innovation and practice of combined models for infectious diseases and help to build smarter and more efficient infectious disease early warning and prediction methods and systems.

Objective:To analyze the clinical characteristics, RAN-binding protein 2 ( RANBP2) gene variations, and prognosis in Chinese acute necrotizing encephalopathy (ANE) patients. Methods:A retrospective analysis of ANE cases registered in the Peking Union Medical College Hospital Encephalitis Registry System from 2022 to 2024, involving patients from Peking Union Medical College Hospital and other hospitals, was conducted. A descriptive study was performed on the clinical characteristics, treatments and prognosis, cerebrospinal fluid examination results, and imaging findings of these patients based on adjusted ANE diagnostic criteria. Whole-exome sequencing technology was used to detect gene mutations in these patients.Results:A total of 18 ANE cases were included, ranged in age from 2 to 72 [20(5, 43)] years. The male-to-female ratio was 4∶5. All patients were found with precipitating infections including COVID-19, influenza A virus and Mycoplasma pneumoniae infections. All patients presented with fever, with varying degrees of consciousness disturbance observed in 16 cases, and seizures in 10 cases. All patients underwent lumbar puncture, with normal or mildly elevated white cell counts [3(2, 13)×10 6/L] and mildly to moderately elevated protein levels [1.90(0.92, 4.65) g/L]. A total of 6 patients were found with extremely elevated interleukin-6 level [950(164, 2 000) pg/ml] in cerebrospinal fluid. Bilateral symmetric thalamic lesions were typical imaging features of ANE, while involvement of other areas such as cortical and subcortical white matter, brainstem, and cerebellum was also observed. A total of 14 patients performed genetic tests while 4 patients were identified with RANBP2 gene mutations (c.1754C>T in 3 cases, c.1966A>G in 1 case). All patients received immunotherapy, and 7 patients died at discharge while other patients presented with neurological sequelae of varying degrees. Conclusions:ANE is a rare and severe parainfectious encephalopathy that can occur in both children and adults. Clinically, it is characterized by rapidly progressing encephalopathy following systematic infection, with bilateral symmetric thalamic lesions. The detection of RANBP2 gene mutations could help make the diagnosis.

Objective:To analyze the efficacy of laparoscopic partial splenectomy and devascularization (LPSPD), open partial splenectomy and devascularization (OPSPD), laparoscopic total splenectomy and devascularization (LTSPD), and open total splenectomy and devascularization (OTSPD) in the treatment of portal hypertension.Methods:Clinical data of 128 patients with portal hypertension admitted to the Department of Hepatobiliary and Pancreatic Surgery of the Affiliated Hospital of Qinghai University and the General Surgery Department of the Fifth Medical Center of the Chinese PLA General Hospital from May 2018 to May 2022 were retrospectively analyzed, including 69 males and 59 females, aged 51.0±10.9 years. According to different surgical methods, patients were divided into four groups, namely the LPSPD group ( n=30), the OPSPD group ( n=32), the LTSPD group ( n=31), and the OTSPD group ( n=35). Operation time, postoperative hospitalization time, postoperative complications such as newly occurred PVT, etc of the four groups of patients were compared. Results:The operation times of the LPSPD, OPSPD, LTSPD and OTSPD group were 240.0 (180.0, 399.0), 209.5 (169.0, 259.0), 207.0 (175.3, 297.5) and 200.0 (162.0, 264.0) minutes, respectively. The postoperative hospital stays were 8.0 (7.0, 12.0), 10.0 (8.3, 11.8), 11.0 (9.0, 13.0) and 13.0 (11.0, 15.0) days, respectively. The incidences of newly occurred PVT after surgery were 3.3% (1/30), 3.1% (1/32), 19.4% (6/31), and 20.0% (7/35), respectively. The above indicators were compared among the four groups respectively, and the differences were statistically significant (all P<0.05). The postoperative hospital stay of the LPSPD group was shorter than that of the LTSPD group [8.0 (7.0, 12.0) vs 11.0 (9.0, 13.0), d, Z=-3.93], and the postoperative hospital stay of the OPSPD group was also shorter than that of the OTSPD group [10.0 (8.3, 11.8) vs 13.0 (11.0, 15.0), d, Z=-3.56; all P<0.001]. The incidence of newly occurred PVT after surgery in the partial splenectomy group (LPSPD and OPSPD) was 3.2% (2/62), which was lower than that in the total splenectomy group [LTSPD and OTSPD, 19.7% (13/66), χ2=8.38, P=0.004]. Conclusion:Compared to total splenectomy, partial splenectomy might be favorable for the shortened postoperative hospital stay and lower incidence of newly occurred portal vein thrombosis.

Objective To study the changes in serum autophagy markers in children with retinopathy of prematurity(ROP)and its clinical value.Methods Premature infants who were screened for ROP in our hospital from January 2020 to December 2022 were selected as the study subjects.Those screened out with ROP at 4-6 weeks of birth were assigned to the ROP group and those without ROP to the control group.The levels of serum autophagy markers LC3-Ⅱ,Beclin-1 and P62 were detected at the 3rd day,1st,2nd and 3rd weeks of birth.The two groups were compared in terms of serum autophagy markers.The diagnostic efficacy of serum autophagy markers on ROP was analyzed.Results There was no significant difference in serum LC3-Ⅱ,Beclin-1 and P62 levels between the ROP group and control group at the 3rd day of birth(P>0.05).At the first,second and third weeks of birth,how-ever,the ROP group showed significantly lower levels of serum LC3-Ⅱ and Beclin-1 but higher level of P62 com-pared to the control group(both P<0.05).The levels of serum LC3-Ⅱ,Beclin-1 and P62 at the first,second and third weeks of birth had diagnostic value for ROP.The children in the ROP group who did not receive mechanical ventilation and oxygen inhalation,and did not develop with sepsis and bronchopulmonary dysplasia showed lower serum LC3-Ⅱ and Beclin-1 levels and higher P62 levels at the first,second and third weeks of birth compared to those without the above-mentioned treatment as well as those complications(all P<0.05).In the ROP group,those with severe ROP showed lower serum LC3-Ⅱ and Beclin-1 levels and higher P62 levels at the 3rd day,and 1st,2nd and 3rd week of birth(all P<0.05).Conclusion The levels of serum autophagy markers in children with ROP show significant changes since the first week of birth,so they have diagnostic efficacy for the diseases.

Objective:To investigate the therapeutic effect difference between first-line treatment with epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKI) and chemotherapy in non-small cell lung cancer (NSCLC) patients with epidermal growth factor receptor (EGFR) rare mutation.Methods:A retrospective case-control study was performed. Data of NSCLC patients with rare EGFR mutation who were treated in Shanxi Province Cancer Hospital from January 2013 to October 2019 were retrospectively analyzed. EGFR mutations in living tissues or blood were detected by using amplification refractory mutation system-polymerase chain reaction (ARMS-PCR) before first-line treatment. According to first-line treatment methods,they were divided into EGFR-TKI treatment group and chemotherapy group. Objective remission rate (ORR) and disease control rate (DCR) of both groups were compared. Kaplan-Meier method was used to draw progression-free survival (PFS) and the overall survival (OS) curves. Log-rank test was used for comparison among groups. Single-factor and multi-factor Cox proportional risk models were used to analyze the influencing factors of PFS and OS.Results:A total of 169 patients with EGFR rare mutations were included, and the age [ M (IQR)] was 63 years (12 years); there were 96 cases (56.8%) < 65 years and 73 cases (43.2%) ≥65 years; 70 (41.4%)males and 99 (58.6%) females; 55 cases (32.5%) had EGFR G719X mutation,45 cases (26.6%) had L861Q mutation, 17 cases (10.1%) had S768I mutation, and 52 cases (30.8%) had complex mutation; 55 cases (32.5%) received the first-line chemotherapy and 114 cases (67.5%) received the first-line EGFR-TKI treatment. In the chemotherapy group, ORR was 36.4% (20/55) and DCR was 85.5% (47/55); in EGFR-TKI treatment group, ORR was 72.8% (83/114) and DCR was 90.4% (103/114). The ORR of EGFR-TKI treatment group was higher than that of chemotherapy group ( χ2 = 20.70, P = 0.001), and there was no statistically significant difference in DCR between two groups ( χ2 = 1.76, P = 0.184). Subgroup analysis showed that ORR in EGFR-TKI treatment group with G719X, L861Q and complex mutations was higher than that of the corresponding mutations in chemotherapy group, and the differences were statistically significant (all P < 0.05), while there were no significant differences in DCR among subgroups (all P > 0.05). The median PFS time was 9.7 months (95% CI: 6.0-13.4 months) and 3.8 months (95% CI: 3.1-7.1 months), respectively in EGFR-TKI treatment group and chemotherapy group, and there was a statistically significant difference in PFS between the two groups ( P < 0.001). The median OS time was 25.6 months (95% CI: 18.0-37.9 months) and 31.7 months (95% CI: 18.0-42.8 months), respectively in EGFR-TKI treatment group and chemotherapy group, and there was no statistically significant difference in OS between the two groups ( P = 0.231). Multivariate Cox regression analysis showed that brain metastasis [with vs. without: HR = 2.306, 95% CI: 1.452-3.661, P < 0.001] and the first-line treatment methods (EGFR-TKI vs. chemotherapy: HR = 0.457, 95% CI:0.317-0.658, P < 0.001) were independent influencing factors for PFS of NSCLC patients with EGFR rare mutation; brain metastasis (with vs. without: HR = 2.087, 95% CI: 1.102-3.953, P = 0.024; unknown vs. without: HR = 2.118,95% CI: 1.274-3.520, P = 0.004) were independent influencing factors for OS of NSCLC patients with EGFR rare mutation. Conclusions:Compared with the first-line chemotherapy, EGFR-TKI first-line treatment could improve objective remission and PFS of NSCLC patients with EGFR rare mutation, while no OS benefit is observed.