To predict COVID-19 severity by building a prediction model based on the clinical manifestations and radiomic features of the thymus in COVID-19 patients.
We retrospectively analyzed the clinical and radiological data from 217 confirmed cases of COVID-19 admitted to Xiangyang NO.1 People's Hospital and Jiangsu Hospital of Chinese Medicine from December 2019 to April 2022 (including 118 mild cases and 99 severe cases). The data were split into the training and test sets at a 7:3 ratio. The cases in the training set were compared in terms of clinical data and radiomic parameters of the lasso regression model. Several models for severity prediction were established based on the clinical and radiomic features of the COVID-19 patients. The DeLong test and decision curve analysis (DCA) were used to compare the performances of several models. Finally, the prediction results were verified on the test set.
For the training set, the univariate analysis showed that BMI, diarrhea, thymic steatosis, anorexia, headache, findings on the chest CT scan, platelets, LDH, AST and radiomic features of the thymus were significantly different between the two groups of patients (P < 0.05). The combination model based on the clinical and radiomic features of COVID-19 patients had the highest predictive value for COVID-19 severity [AUC: 0.967 (OR 0.0115, 95%CI: 0.925-0.989)] vs. the clinical feature-based model [AUC: 0.772 (OR 0.0387, 95%CI: 0.697-0.836), P < 0.05], laboratory-based model [AUC: 0.687 (OR 0.0423, 95%CI: 0.608-0.760), P < 0.05] and model based on CT radiomics [AUC: 0.895 (OR 0.0261, 95%CI: 0.835-0.938), P < 0.05]. DCA also confirmed the high clinical net benefits of the combination model. The nomogram drawn based on the combination model could help differentiate between the mild and severe cases of COVID-19 at an early stage. The predictions from different models were verified on the test set.
Severe cases of COVID-19 had a higher level of thymic involution. The thymic differentiation in radiomic features was related to disease progression. The combination model based on the radiomic features of the thymus could better promote early clinical intervention of COVID-19 and increase the cure rate.
Citation: Peng An, Xiumei Li, Ping Qin, YingJian Ye, Junyan Zhang, Hongyan Guo, Peng Duan, Zhibing He, Ping Song, Mingqun Li, Jinsong Wang, Yan Hu, Guoyan Feng, Yong Lin. Predicting model of mild and severe types of COVID-19 patients using Thymus CT radiomics model: A preliminary study[J]. Mathematical Biosciences and Engineering, 2023, 20(4): 6612-6629. doi: 10.3934/mbe.2023284
[1] | Zongying Feng, Guoqiang Tan . Dynamic event-triggered control for neural networks with sensor saturations and stochastic deception attacks. Electronic Research Archive, 2025, 33(3): 1267-1284. doi: 10.3934/era.2025056 |
[2] | Yawei Liu, Guangyin Cui, Chen Gao . Event-triggered synchronization control for neural networks against DoS attacks. Electronic Research Archive, 2025, 33(1): 121-141. doi: 10.3934/era.2025007 |
[3] | Xingyue Liu, Kaibo Shi, Yiqian Tang, Lin Tang, Youhua Wei, Yingjun Han . A novel adaptive event-triggered reliable control approach for networked control systems with actuator faults. Electronic Research Archive, 2023, 31(4): 1840-1862. doi: 10.3934/era.2023095 |
[4] | Chao Ma, Hang Gao, Wei Wu . Adaptive learning nonsynchronous control of nonlinear hidden Markov jump systems with limited mode information. Electronic Research Archive, 2023, 31(11): 6746-6762. doi: 10.3934/era.2023340 |
[5] | Hangyu Hu, Fan Wu, Xiaowei Xie, Qiang Wei, Xuemeng Zhai, Guangmin Hu . Critical node identification in network cascading failure based on load percolation. Electronic Research Archive, 2023, 31(3): 1524-1542. doi: 10.3934/era.2023077 |
[6] | Chengbo Yi, Jiayi Cai, Rui Guo . Synchronization of a class of nonlinear multiple neural networks with delays via a dynamic event-triggered impulsive control strategy. Electronic Research Archive, 2024, 32(7): 4581-4603. doi: 10.3934/era.2024208 |
[7] | Liping Fan, Pengju Yang . Load forecasting of microgrid based on an adaptive cuckoo search optimization improved neural network. Electronic Research Archive, 2024, 32(11): 6364-6378. doi: 10.3934/era.2024296 |
[8] | Ramalingam Sakthivel, Palanisamy Selvaraj, Oh-Min Kwon, Seong-Gon Choi, Rathinasamy Sakthivel . Robust memory control design for semi-Markovian jump systems with cyber attacks. Electronic Research Archive, 2023, 31(12): 7496-7510. doi: 10.3934/era.2023378 |
[9] | Nacera Mazouz, Ahmed Bengermikh, Abdelhamid Midoun . Dynamic design and optimization of a power system DC/DC converter using peak current mode control. Electronic Research Archive, 2025, 33(4): 1968-1997. doi: 10.3934/era.2025088 |
[10] | Yejin Yang, Miao Ye, Qiuxiang Jiang, Peng Wen . A novel node selection method for wireless distributed edge storage based on SDN and a maldistributed decision model. Electronic Research Archive, 2024, 32(2): 1160-1190. doi: 10.3934/era.2024056 |
To predict COVID-19 severity by building a prediction model based on the clinical manifestations and radiomic features of the thymus in COVID-19 patients.
We retrospectively analyzed the clinical and radiological data from 217 confirmed cases of COVID-19 admitted to Xiangyang NO.1 People's Hospital and Jiangsu Hospital of Chinese Medicine from December 2019 to April 2022 (including 118 mild cases and 99 severe cases). The data were split into the training and test sets at a 7:3 ratio. The cases in the training set were compared in terms of clinical data and radiomic parameters of the lasso regression model. Several models for severity prediction were established based on the clinical and radiomic features of the COVID-19 patients. The DeLong test and decision curve analysis (DCA) were used to compare the performances of several models. Finally, the prediction results were verified on the test set.
For the training set, the univariate analysis showed that BMI, diarrhea, thymic steatosis, anorexia, headache, findings on the chest CT scan, platelets, LDH, AST and radiomic features of the thymus were significantly different between the two groups of patients (P < 0.05). The combination model based on the clinical and radiomic features of COVID-19 patients had the highest predictive value for COVID-19 severity [AUC: 0.967 (OR 0.0115, 95%CI: 0.925-0.989)] vs. the clinical feature-based model [AUC: 0.772 (OR 0.0387, 95%CI: 0.697-0.836), P < 0.05], laboratory-based model [AUC: 0.687 (OR 0.0423, 95%CI: 0.608-0.760), P < 0.05] and model based on CT radiomics [AUC: 0.895 (OR 0.0261, 95%CI: 0.835-0.938), P < 0.05]. DCA also confirmed the high clinical net benefits of the combination model. The nomogram drawn based on the combination model could help differentiate between the mild and severe cases of COVID-19 at an early stage. The predictions from different models were verified on the test set.
Severe cases of COVID-19 had a higher level of thymic involution. The thymic differentiation in radiomic features was related to disease progression. The combination model based on the radiomic features of the thymus could better promote early clinical intervention of COVID-19 and increase the cure rate.
Gender hormones regulate structure and function of many tissue and organ systems [1],[2]. Sexual dimorphism is defined as “the differences in appearance between males and females of the same species, such as in colour, shape, size, and structure, that are caused by the inheritance of one or the other sexual pattern in the genetic material” [3]. Some studies have reported that gender hormones affect renal morphology and physiology, and gender differences exist in the prevalence and prognosis of renal diseases. However, there are inconsistent results across study outcomes. There are also limited data available on this issue in humans [1],[2],[4],[5].
It is emphasizes that women have a slower rate of decline in renal function than men. This condition can be due to gender differences in kidney size and weight, biological, metabolic and hemodynamic processes [1],[4]. In a study of 13,925 Chinese adults, Xu et al. [6] reported that the rates of decline in estimated glomerular filtration rate in men in both the at-risk group and the chronic kidney disease (CKD) group were faster compared to women, after referencing to the healthy group. Fanelli et al. [7] investigated gender differences in the progression of experimental CKD induced by chronic nitric oxide inhibition in rats. Their findings have indicated that female rats developed less severe CKD compared to males. According to Fanelli et al. [7], “female renoprotection could be promoted by both the estrogen anti-inflammatory activity and/or by the lack of testosterone, related to renin-angiotensin-aldosterone system hyperactivation and fibrogenesis” [p. 1]. Other studies have also reported that CKD was slightly more common among women than men [8],[9]. In a prospective, community-based, cohort study of 5488 participants from the Netherlands, Halbesma et al. [10] investigated gender differences as predictors of the decline of renal function. They found that systolic blood pressure and plasma glucose level negatively associated with renal function decline for both genders. Interestingly, this follow-up study demonstrated that waist circumference was positively associated with renal function in men only [10]. In another community-based, cohort study of 1876 Japanese adults, a higher body mass index was also found to be an independent risk factor for the development of CKD in women only [11]. On the other hand, compared with men, women tend to initiate hemodialysis with an arteriovenous fistula less frequently, and have greater risk of arteriovenous fistula failure [8]. Carrero et al. [5] also reported that women are less likely to receive kidney transplants than men. Further research is therefore needed to better understand the effect of gender on kidney function and health outcomes.
[1] |
M. Herrero-Montes, C. Fernández-de-Las-Peñas, D. Ferrer-Pargada, S. Tello-Mena, I. Cancela-Cilleruelo, J. Rodríguez-Jiménez, et al., Prevalence of neuropathic component in post-COVID pain symptoms in previously hospitalized COVID-19 survivors, Int. J. Clin. Pract., 2022 (2022), 3532917. https://doi:10.1155/2022/3532917 doi: 10.1155/2022/3532917
![]() |
[2] |
S. Abuhammad, O. F. Khabour, K. H. Alzoubi, F. El-Zubi, S. H. Hamaieh, Respiratory infectious diseases and adherence to nonpharmacological interventions for overcoming COVID-19 pandemic: A self-reported study, Int. J. Clin. Pract., 2022 (2022), 4495806. https://doi:10.1155/2022/4495806 doi: 10.1155/2022/4495806
![]() |
[3] |
N. Demir, B. Yüzbasıoglu, T. Calhan, S. Ozturk, Prevalence and prognostic importance of high fibrosis-4 index in COVID-19 patients, Int. J. Clin. Pract., 2022 (2022), 1734896. https://doi:10.1155/2022/1734896 doi: 10.1155/2022/1734896
![]() |
[4] |
S. Tharwat, H. A. Abdelsalam, A. Abdelsalam, M. K. Nassar, COVID-19 vaccination intention and vaccine hesitancy among patients with autoimmune and autoinflammatory rheumatological diseases: A survey, Int. J. Clin. Pract., 2022 (2022), 5931506. https://doi:10.1155/2022/5931506 doi: 10.1155/2022/5931506
![]() |
[5] |
Y. Liu, Y. Pan, Z. Hu, M. Wu, C. Wang, Z. Feng, et al., Thymosin Alpha 1 reduces the mortality of severe coronavirus disease 2019 by restoration of lymphocytopenia and reversion of exhausted T cells, Clin. Infect. Dis., 71 (2020), 2150–2157. https://doi:10.1093/cid/ciaa630 doi: 10.1093/cid/ciaa630
![]() |
[6] |
V. J. Sharmila, D. Jemi Florinabel, Deep learning algorithm for COVID-19 classification using chest X-ray images, Comput. Math. Methods Med., 2021 (2021), 9269173. https://doi:10.1155/2021/9269173 doi: 10.1155/2021/9269173
![]() |
[7] |
W. C. Serena Low, J. H. Chuah, C. A. T. H. Tee, S. Anis, M. A. Shoaib, A. Faisal, et al., An overview of deep learning techniques on chest X-ray and CT scan identification of COVID-19, Comput. Math. Methods Med., 2021 (2021), 5528144. https://doi:10.1155/2021/5528144 doi: 10.1155/2021/5528144
![]() |
[8] |
M. Nakhaeizadeh, M. Chegeni, M. Adhami, H. Sharifi, M. A. Gohari, A. Iranpour, et al., Estimating the number of COVID-19 cases and impact of new COVID-19 variants and vaccination on the population in Kerman, Iran: A mathematical modeling study, Comput. Math. Methods Med., 2022 (2022), 6624471. https://doi:10.1155/2022/6624471 doi: 10.1155/2022/6624471
![]() |
[9] |
J. B. Ackman, B. Kovacina, B. W. Carter, C. C. Wu, A. Sharma, J. O. Shepard, et al., Sex difference in normal thymic appearance in adults 20–30 years of age, Radiology, 268 (2013), 245–253. https://doi:10.1148/radiol.13121104 doi: 10.1148/radiol.13121104
![]() |
[10] |
M. Takesh, S. Adams, Imaging comparison between (18) F-FDG-PET/CT and (18) F-Flouroethyl choline PET/CT in rare case of Thymus Carcinoma exhibiting a positive choline uptake, Case Rep. Oncol. Med., 2013 (2013), 464396. https://doi:10.1155/2013/464396 doi: 10.1155/2013/464396
![]() |
[11] |
N. Simanovsky, N. Hiller, N. Loubashevsky, K. Rozovsky, Normal CT characteristics of the thymus in adults, Eur. J. Radiol., 81 (2012), 3581–3586. https://doi:10.1016/j.ejrad.2011.12.015 doi: 10.1016/j.ejrad.2011.12.015
![]() |
[12] |
T. Araki, M. Nishino, W. Gao, J. Dupuis, G. M. Hunninghake, T. Murakami, et al., Normal thymus in adults: appearance on CT and associations with age, sex, BMI and smoking, Eur. Radiol., 26 (2016), 15–24. https://doi:10.1007/s00330-015-3796-y doi: 10.1007/s00330-015-3796-y
![]() |
[13] |
H. Zhou, R. Xu, H. Mei, L. Zhang, Q. Yu, R. Liu, et al., Application of enhanced T1WI of MRI Radiomics in Glioma grading, Int. J. Clin. Pract., 2022 (2022), 3252574. https://doi:10.1155/2022/3252574 doi: 10.1155/2022/3252574
![]() |
[14] |
J. Wang, J. Zeng, H. Li, X. Yu, A deep learning radiomics analysis for survival prediction in Esophageal cancer, J. Healthcare Eng., 2022 (2022), 4034404. https://doi:10.1155/2022/4034404 doi: 10.1155/2022/4034404
![]() |
[15] |
Y. Wang, G. Feng, J. Wang, P. An, P. Duan, Y. Hu, et al., Contrast-enhanced ultrasound-magnetic resonance imaging radiomics based model for predicting the biochemical recurrence of prostate cancer: A feasibility study, Comput. Math. Methods Med., 2022 (2022), 8090529. https://doi:10.1155/2022/8090529 doi: 10.1155/2022/8090529
![]() |
[16] |
I. Malinauskaite, J. Hofmeister, S. Burgermeister, A. Neroladaki, M. Hamard, X. Montet, et al., Radiomics and machine learning differentiate soft-tissue lipoma and liposarcoma better than musculoskeletal radiologists, Sarcoma, 2020 (2022), 7163453. https://doi:10.1155/2020/7163453 doi: 10.1155/2020/7163453
![]() |
[17] |
P. An, J. Zhang, Y. Li, P. Duan, Y. Hu, X. Li, et al., Clinical and imaging data-based model for predicting Reversible Posterior Leukoencephalopathy Syndrome (RPLS) in pregnant women with severe preeclampsia or eclampsia and analysis of perinatal outcomes, Int. J. Clin. Pract., 2022 (2022), 6990974. https://doi:10.1155/2022/6990974 doi: 10.1155/2022/6990974
![]() |
[18] |
P. An, J. Zhang, F. Yang, Z. Wang, Y. Hu, X. Li, USMRI features and clinical data-based model for predicting the degree of placenta accreta spectrum disorders and developing prediction models, Int. J. Clin. Pract., 2022 (2022), 9527412. https://doi:10.1155/2022/9527412 doi: 10.1155/2022/9527412
![]() |
[19] |
P. An, W. Gu, S. Luo, M. Zhang, Y. Wang, Q. X. Li, Radiological changes on chest CT following COVID-19 infection, Ann. Acad. Med. Singapore, 50 (2021), 346–348. https://doi:10.47102/annals-acadmedsg.2020208 doi: 10.47102/annals-acadmedsg.2020208
![]() |
[20] |
P. An, P. Song, Y. Wang, B. Liu, Asymptomatic patients with novel coronavirus disease (COVID-19), Balkan Med. J., 37 (2020), 229–230. https://doi:10.4274/balkanmedj.galenos.2020.2020.4.20 doi: 10.4274/balkanmedj.galenos.2020.2020.4.20
![]() |
[21] |
P. An, P. Song, K. Lian, Y. Wang, CT manifestations of novel coronavirus pneumonia: A case report, Balkan Med. J., 37 (2020), 163–165. https://doi:10.4274/balkanmedj.galenos.2020.2020.2.15 doi: 10.4274/balkanmedj.galenos.2020.2020.2.15
![]() |
[22] |
P. An, B. J. Wood, W. Li, M. Zhang, Y. Ye, Postpartum exacerbation of antenatal COVID-19 pneumonia in 3 women, CMAJ, 192 (2020), E603–E606. https://doi:10.1503/cmaj.200553 doi: 10.1503/cmaj.200553
![]() |
[23] |
P. An, Y. Ye, M. Chen, Y. Chen, W. Fan, Y. Wang, Management strategy of novel coronavirus (COVID-19) pneumonia in the radiology department: a Chinese experience, Diagn. Interv. Radiol., 26 (2020), 200–203. https://doi:10.5152/dir.2020.20167 doi: 10.5152/dir.2020.20167
![]() |
[24] |
C. Kellogg, O. Equils, The role of the thymus in COVID-19 disease severity: implications for antibody treatment and immunization, Hum. Vaccines Immunother., 17 (2021), 638–643. https://doi:10.1080/21645515.2020.1818519 doi: 10.1080/21645515.2020.1818519
![]() |
[25] |
P. Cuvelier, H. Roux, A. Couëdel-Courteille, J. Dutrieux, C. Naudin, B. C. de Muylder, et al., Protective reactive thymus hyperplasia in COVID-19 acute respiratory distress syndrome, Crit. Care., 25 (2021), 4. https://doi:10.1186/s13054-020-03440-1 doi: 10.1186/s13054-020-03440-1
![]() |
[26] |
R. Thomas, W. Wang, D. M. Su, Contributions of age-related thymic involution to immunosenescence and inflammaging, Immun. Ageing, 17 (2020), 2. https://doi:10.1186/s12979-020-0173-8 doi: 10.1186/s12979-020-0173-8
![]() |
[27] |
W. Wang, R. Thomas, J. Oh, D. M. Su, Thymic aging may be associated with COVID-19 pathophysiology in the elderly, Cells, 10 (2021), 628. https://doi:10.3390/cells10030628 doi: 10.3390/cells10030628
![]() |
[28] |
S. Rehman, T. Majeed, M. A. Ansari, U. Ali, H. Sabit, E. A. Al-Suhaimi, Current scenario of COVID-19 in pediatric age group and physiology of immune and thymus response, Saudi J. Biol. Sci., 27 (2020), 2567–2573. https://doi:10.1016/j.sjbs.2020.05.024 doi: 10.1016/j.sjbs.2020.05.024
![]() |
[29] |
K. A. Harrington, D. S. Kennedy, B. Tang, C. Hickie, E. Phelan, W. Torreggiani, et al., Computed tomographic evaluation of the thymus-does obesity affect thymic fatty involution in a healthy young adult population, Br. J. Radiol., 91 (2018), 20170609. https://doi:10.1259/bjr.20170609 doi: 10.1259/bjr.20170609
![]() |
[30] |
F. Nasseri, F. Eftekhari, Clinical and radiologic review of the normal and abnormal thymus: pearls and pitfalls, Radiographics, 30 (2010), 413–428. https://doi:10.1148/rg.302095131 doi: 10.1148/rg.302095131
![]() |
[31] |
J. L. Zhang, Y. H. Li, L. L. Wang, H. Q. Liu, S. Y. Lu, Y. Liu, et al., Azvudine is a thymus-homing anti-SARS-CoV-2 drug effective in treating COVID-19 patients, Signal Transduction Targeted Ther., 6 (2021), 414. https://doi:10.1038/s41392-021-00835-6 doi: 10.1038/s41392-021-00835-6
![]() |
[32] |
M. E. Mayerhoefer, A. Materka, G. Langs, I. Häggström, P. Szczypiński, P. Gibbs, et al., Introduction to radiomics, J. Nucl. Med., 61 (2020), 488–495. https://doi:10.2967/jnumed.118.222893 doi: 10.2967/jnumed.118.222893
![]() |
[33] |
M. R. Chetan, F. V. Gleeson, Radiomics in predicting treatment response in non-small-cell lung cancer: current status, challenges and future perspectives, Eur. Radiol., 31 (2021), 1049–1058. https://doi:10.1007/s00330-020-07141-9 doi: 10.1007/s00330-020-07141-9
![]() |
[34] |
M. Seyit, E. Avci, A. Yilmaz, H. Senol, M. Ozen, A. Oskay, Predictive values of coagulation parameters to monitor COVID-19 patients, Int. J. Clin. Pract., 2022 (2022), 8436248. https://doi:10.1155/2022/8436248 doi: 10.1155/2022/8436248
![]() |
[35] |
I. Tsougos, A. Vamvakas, C. Kappas, I. Fezoulidis, K. Vassiou, Application of radiomics and decision support systems for breast mr differential diagnosis, Comput. Math. Methods Med., 2018 (2018), 7417126. https://doi:10.1155/2018/7417126 doi: 10.1155/2018/7417126
![]() |
[36] |
M. Umesh Pai, A. A. Ardakani, A. Kamath, U. Raghavendra, A. Gudigar, N. Venkatesh, et al., Novel radiomics features for automated detection of cardiac abnormality in patients with pacemaker, Comput. Math. Methods Med., 2022 (2022), 1279749. https://doi:10.1155/2022/1279749 doi: 10.1155/2022/1279749
![]() |
[37] |
S. A. Harmon, T. H. Sanford, S. Xu, E. B. Turkbey, H. Roth, Z. Xu, et al., Artificial intelligence for the detection of COVID-19 pneumonia on chest CT using multinational datasets, Nat. Commun., 11 (2020), 4080. https://doi:10.1038/s41467-020-17971-2 doi: 10.1038/s41467-020-17971-2
![]() |
[38] |
S. Cournane, R. Conway, D. Byrne, D. O'Riordan, B. Silke, Predicting outcomes in emergency medical admissions using a laboratory only nomogram, Comput. Math. Methods Med., 2017 (2017), 5267864. https://doi:10.1155/2017/5267864 doi: 10.1155/2017/5267864
![]() |
[39] |
S. Tian, Y. Guo, J. Fu, Z. Li, J. Li, X. Tian, Prognostic value of immunotyping combined with targeted therapy in patients with non-small-cell lung cancer and establishment of nomogram model, Comput. Math. Methods Med., 2022 (2022), 3049619. https://doi:10.1155/2022/3049619 doi: 10.1155/2022/3049619
![]() |
[40] |
R. Qin, H. Zhang, L. Jiang, K. Qiao, J. Hai, J. Chen, et al., Multicenter computer-aided diagnosis for lymph nodes using unsupervised domain-adaptation networks based on cross-domain confounding representations, Comput. Math. Methods Med., 2020 (2020), 3709873. https://doi:10.1155/2020/3709873 doi: 10.1155/2020/3709873
![]() |
[41] |
M. Brambilla, R. Matheoud, C. Basile, C. Bracco, I. Castiglioni, C. Cavedon, et al., An adaptive thresholding method for BTV estimation incorporating PET reconstruction parameters: A multicenter study of the robustness and the reliability, Comput. Math. Methods Med., 2015 (2015), 571473. https://doi:10.1155/2015/571473 doi: 10.1155/2015/571473
![]() |