Skip to content
2000
Volume 21, Issue 1
  • ISSN: 1573-4056
  • E-ISSN: 1875-6603
file format text download EPUB
file format pdf download PDF
side by side viewer icon HTML

Abstract

Introduction

This study aims to improve the accuracy of distinguishing Tuberculous Spondylitis (TBS) from Brucella Spondylitis (BS) by developing radiomics models using Deep Learning and CT images enhanced with Super-Resolution (SR).

Methods

A total of 94 patients diagnosed with BS or TBS were randomly divided into training (n=65) and validation (n=29) groups in a 7:3 ratio. In the training set, there were 40 BS and 25 TBS patients, with a mean age of 58.34 ± 12.53 years. In the validation set, there were 17 BS and 12 TBS patients, with a mean age of 58.48 ± 12.29 years. Standard CT images were enhanced using SR, improving spatial resolution and image quality. The lesion regions (ROIs) were manually segmented, and radiomics features were extracted. ResNet18 and ResNet34 were used for deep learning feature extraction and model training. Four multi-layer perceptron (MLP) models were developed: clinical, radiomics (Rad), deep learning (DL), and a combined model. Model performance was assessed using five-fold cross-validation, ROC, and decision curve analysis (DCA).

Results

Statistical significance was assessed, with key clinical and imaging features showing significant differences between TBS and BS (, gender, p=0.0038; parrot beak appearance, p<0.001; dead bone, p<0.001; deformities of the spinal posterior process, p=0.0044; psoas abscess, p<0.001). The combined model outperformed others, achieving the highest AUC (0.952), with ResNet34 and SR-enhanced images further boosting performance. Sensitivity reached 0.909, and Specificity was 0.941. DCA confirmed clinical applicability.

Discussion

The integration of SR-enhanced CT imaging and deep learning radiomics appears to improve diagnostic differentiation between BS and TBS. The combined model, especially when using ResNet34 and GAN-based super-resolution, demonstrated better predictive performance. High-resolution imaging may facilitate better lesion delineation and more robust feature extraction. Nevertheless, further validation with larger, multicenter cohorts is needed to confirm generalizability and reduce potential bias from retrospective design and imaging heterogeneity.

Conclusion

This study suggests that integrating Deep Learning Radiomics with Super-Resolution may improve the differentiation between TBS and BS compared to standard CT imaging. However, prospective multi-center studies are necessary to validate its clinical applicability.

© 2025 The Author(s). Published by Bentham Science Publishers.
This is an open access article published under CC BY 4.0 https://creativecommons.org/licenses/by/4.0/legalcode
Loading

Article metrics loading...

/content/journals/cmir/10.2174/0115734056380084250720064859
2025-08-04
2025-09-19

  • From This Site

    /content/journals/cmir/10.2174/0115734056380084250720064859
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

/deliver/fulltext/cmir/21/1/CMIR-21-E15734056380084.html?itemId=/content/journals/cmir/10.2174/0115734056380084250720064859&mimeType=html&fmt=ahah

References

  1. PaiM. KasaevaT. SwaminathanS. Covid-19’s devastating effect on tuberculosis care - A path to recovery.N. Engl. J. Med.2022386161490149310.1056/NEJMp211814534986295
     [Google Scholar]
  2. SoleraJ. LozanoE. Martínez-AlfaroE. EspinosaA. CastillejosM.L. AbadL. Brucellar spondylitis: Review of 35 cases and literature survey.Clin. Infect. Dis.19992961440144910.1086/31352410585793
     [Google Scholar]
  3. Sanchez-GonzalezJ. Garcia-DelangeT. MartosF. ColmeneroJ.D. Thrombosis of the abdominal aorta secondary to Brucella spondylitis.Infection199624326126210.1007/BF017811088811368
     [Google Scholar]
  4. YeeN. RoachD.J. Infected abdominal aortic aneurysm caused by spinal brucellar infection.AJR Am. J. Roentgenol.199616741068106910.2214/ajr.167.4.88194208819420
     [Google Scholar]
  5. MuehlematterU.J. MannilM. BeckerA.S. VokingerK.N. FinkenstaedtT. OsterhoffG. FischerM.A. GuggenbergerR. Vertebral body insufficiency fractures: Detection of vertebrae at risk on standard CT images using texture analysis and machine learning.Eur. Radiol.20192952207221710.1007/s00330‑018‑5846‑830519934
     [Google Scholar]
  6. LiuX. LiH. JinC. NiuG. GuoB. ChenY. YangJ. Differentiation between brucellar and tuberculous spondylodiscitis in the acute and subacute stages by MRI.Acad. Radiol.20182591183118910.1016/j.acra.2018.01.02829609954
     [Google Scholar]
  7. BousisD. The role of deep learning in diagnosing colorectal cancer.Prz. Gastroenterol.202318326627310.5114/pg.2023.129494
     [Google Scholar]
  8. ChlorogiannisD.D. Tissue classification and diagnosis of colorectal cancer histopathology images using deep learning algorithms. Is the time ripe for clinical practice implementation?Prz. Gastroenterol.202318435336710.5114/pg.2023.130337
     [Google Scholar]
  9. DimopoulosP. The role of artificial intelligence and image processing in the diagnosis, treatment, and prognosis of liver cancer: A narrative review.Prz. Gastroenterol.202419322123010.5114/pg.2024.143147
     [Google Scholar]
  10. HuangY. ZhuT. ZhangX. LiW. ZhengX. ChengM. JiF. ZhangL. YangC. WuZ. YeG. LinY. WangK. Longitudinal MRI-based fusion novel model predicts pathological complete response in breast cancer treated with neoadjuvant chemotherapy: A multicenter, retrospective study.EClinicalMedicine20235810189910.1016/j.eclinm.2023.10189937007742
     [Google Scholar]
  11. WangH. WangL. LeeE.H. ZhengJ. ZhangW. HalabiS. LiuC. DengK. SongJ. YeomK.W. Decoding COVID-19 pneumonia: Comparison of deep learning and radiomics CT image signatures.Eur. J. Nucl. Med. Mol. Imaging20214851478148610.1007/s00259‑020‑05075‑433094432
     [Google Scholar]
  12. KrizhevskyA. SutskeverI. HintonG.E. ImageNet classification with deep convolutional neural networks.Commun. ACM2017606849010.1145/3065386
     [Google Scholar]
  13. de FariasE.C. di NoiaC. HanC. SalaE. CastelliM. RundoL. Impact of GAN-based lesion-focused medical image super-resolution on the robustness of radiomic features.Sci. Rep.20211112136110.1038/s41598‑021‑00898‑z34725417
     [Google Scholar]
  14. ChiJ. SunZ. WangH. LyuP. YuX. WuC. CT image super-resolution reconstruction based on global hybrid attention.Comput. Biol. Med.202215010611210.1016/j.compbiomed.2022.10611236209555
     [Google Scholar]
  15. Abstracts of the 4th European Congress of Chemotherapy and InfectionParis, France, 4-7 May 2002.
     [Google Scholar]
  16. HuangY.H. ShiZ.Y. ZhuT. ZhouT.H. LiY. LiW. QiuH. WangS.Q. HeL.F. WuZ.Y. LinY. WangQ. GuW.C. GuC.C. SongX.Y. ZhouY. GuanD.G. WangK. Longitudinal MRI-driven multi-modality approach for predicting pathological complete response and B cell infiltration in breast cancer.Adv. Sci.20251212241370210.1002/advs.20241370239921294
     [Google Scholar]
  17. SeoniS. ShahiniA. MeiburgerK.M. MarzolaF. RotunnoG. AcharyaU.R. MolinariF. SalviM. All you need is data preparation: A systematic review of image harmonization techniques in Multi-center/device studies for medical support systems.Comput. Methods Programs Biomed.202425010820010.1016/j.cmpb.2024.10820038677080
     [Google Scholar]
  18. WolterinkJ.M. Generative adversarial networks: A primer for radiologists.Radiographics202141384085710.1148/rg.2021200151
     [Google Scholar]
  19. BejaniM.M. GhateeM. A systematic review on overfitting control in shallow and deep neural networks.Artif. Intell. Rev.20215486391643810.1007/s10462‑021‑09975‑1
     [Google Scholar]
  20. JohnsonW.E. LiC. RabinovicA. Adjusting batch effects in microarray expression data using empirical Bayes methods.Biostatistics20078111812710.1093/biostatistics/kxj03716632515
     [Google Scholar]
  21. LeithnerD. SchöderH. HaugA. VargasH.A. GibbsP. HäggströmI. RauschI. WeberM. BeckerA.S. SchwartzJ. MayerhoeferM.E. Impact of ComBat harmonization on PET radiomics-based tissue classification: A dual-center PET/MRI and PET/CT study.J. Nucl. Med.202263101611161610.2967/jnumed.121.26310235210300
     [Google Scholar]
  22. ShenW. SongZ. ZhongX. HuangM. ShenD. GaoP. QianX. WangM. HeX. WangT. LiS. SongX. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform.iMeta202213e3610.1002/imt2.3638868713
     [Google Scholar]
  23. TaminauJ. MeganckS. LazarC. SteenhoffD. ColettaA. MolterC. DuqueR. de SchaetzenV. Weiss SolísD.Y. BersiniH. NowéA. Unlocking the potential of publicly available microarray data using in Silico Db and in Silico Merging R/Bioconductor packages.BMC Bioinformatics2012Dec241333510.1186/1471‑2105‑13‑335
     [Google Scholar]
  24. Hanchuan Peng Fuhui Long DingC. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy.IEEE Trans. Pattern Anal. Mach. Intell.20052781226123810.1109/TPAMI.2005.15916119262
     [Google Scholar]
  25. ZhangJ. LiuJ. LiangZ. XiaL. ZhangW. XingY. ZhangX. TangG. Differentiation of acute and chronic vertebral compression fractures using conventional CT based on deep transfer learning features and hand-crafted radiomics features.BMC Musculoskelet. Disord.202324116510.1186/s12891‑023‑06281‑536879285
     [Google Scholar]
  26. RastegarS. VaziriM. QasempourY. AkhashM.R. AbdalvandN. ShiriI. AbdollahiH. ZaidiH. Radiomics for classification of bone mineral loss: A machine learning study.Diagn. Interv. Imaging2020101959961010.1016/j.diii.2020.01.00832033913
     [Google Scholar]
  27. AkansuA.N. SerdijnW.A. SelesnickI.W. Emerging applications of wavelets: A review.Phys. Commun.20103111810.1016/j.phycom.2009.07.001
     [Google Scholar]
  28. HeK.Z. Xiangyu and Ren, Shaoqing and Sun, Jian, Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)201677077810.1109/CVPR.2016.90
     [Google Scholar]
  29. XiaoB. SunH. MengY. PengY. YangX. ChenS. YanZ. ZhengJ. Classification of microcalcification clusters in digital breast tomosynthesis using ensemble convolutional neural network.Biomed. Eng. Online20212017110.1186/s12938‑021‑00908‑134320986
     [Google Scholar]
  30. HalliganS. MenuY. MallettS. Why did European radiology reject my radiomic biomarker paper? how to correctly evaluate imaging biomarkers in a clinical setting.Eur. Radiol.202131129361936810.1007/s00330‑021‑07971‑134003349
     [Google Scholar]
  31. XuH. AbdallahN. MarionJ.M. ChauvetP. TauberC. CarlierT. LuL. HattM. Radiomics prognostic analysis of PET/CT images in a multicenter head and neck cancer cohort: Investigating ComBat strategies, sub-volume characterization, and automatic segmentation.Eur. J. Nucl. Med. Mol. Imaging20235061720173410.1007/s00259‑023‑06118‑236690882
     [Google Scholar]
  32. JiaW. Investigation of ComBat harmonization on radiomic and deep features from multi-center abdominal MRI data.J. Imaging Inform. Med.20253821016102710.1007/s10278‑024‑01253‑0
     [Google Scholar]
  33. ZhouC. ZhouJ. LvY. BatuerM. HuangJ. ZhongJ. ZhongH. QinG. The impact of the novel CovBat harmonization method on enhancing radiomics feature stability and machine learning model performance: A multi-center, multi-device study.Eur. J. Radiol.202518411195610.1016/j.ejrad.2025.11195639908939
     [Google Scholar]
  34. WangW. FanZ. ZhenJ. MRI radiomics-based evaluation of tuberculous and brucella spondylitis.J. Int. Med. Res.20235180300060523119515610.1177/0300060523119515637656968
     [Google Scholar]
  35. ZhangX. DongX. SaripanM.I. DuD. WuY. WangZ. CaoZ. WenD. LiuY. MarhabanM.H. Deep learning PET / CT -based radiomics integrates clinical data: A feasibility study to distinguish between tuberculosis nodules and lung cancer.Thorac. Cancer202314191802181110.1111/1759‑7714.1492437183577
     [Google Scholar]
  36. HuangY. YaoZ. LiL. MaoR. HuangW. HuZ. HuY. WangY. GuoR. TangX. YangL. WangY. LuoR. YuJ. ZhouJ. Deep learning radiopathomics based on preoperative US images and biopsy whole slide images can distinguish between luminal and non-luminal tumors in early-stage breast cancers.EBioMedicine20239410470610.1016/j.ebiom.2023.10470637478528
     [Google Scholar]
  37. GuerreiroJ. TomásP. GarciaN. AidosH. Super-resolution of magnetic resonance images using generative adversarial networks.Comput. Med. Imaging Graph.202310810228010.1016/j.compmedimag.2023.10228037597380
     [Google Scholar]
  38. XieH. ZhangT. SongW. WangS. ZhuH. ZhangR. ZhangW. YuY. ZhaoY. Super-resolution of Pneumocystis carinii pneumonia CT via self-attention GAN.Comput. Methods Programs Biomed.202121210646710.1016/j.cmpb.2021.10646734715519
     [Google Scholar]
  39. HouM. ZhouL. SunJ. Deep-learning-based 3D super-resolution MRI radiomics model: Superior predictive performance in preoperative T-staging of rectal cancer.Eur. Radiol.202233111010.1007/s00330‑022‑08952‑835726100
     [Google Scholar]
  40. LiC. HeZ. LvF. LiaoH. XiaoZ. Predicting the prognosis of HIFU ablation of uterine fibroids using a deep learning-based 3D super-resolution DWI radiomics model: A multicenter study.Acad. Radiol.202431124996500710.1016/j.acra.2024.06.02738969576
     [Google Scholar]
  41. MarquesG. Enhanced Telemedicine and e-Health.Springer202110.1007/978‑3‑030‑70111‑6
     [Google Scholar]
  42. SaleemT.J. ChishtiM.A. Deep learning for internet of things data analytics.Procedia Comput. Sci.201916338139010.1016/j.procs.2019.12.120
     [Google Scholar]
  43. MulitaF. VerrasG.I. AnagnostopoulosC.N. KotisK. A smarter health through the internet of surgical things.Sensors20222212457710.3390/s2212457735746359
     [Google Scholar]
  44. BandS.S. ArdabiliS. YarahmadiA. PahlevanzadehB. KianiA.K. BeheshtiA. Alinejad-RoknyH. DehzangiI. ChangA. MosaviA. MoslehpourM. A survey on machine learning and internet of medical things-based approaches for handling covid-19: Meta-analysis.Front. Public Health20221086923810.3389/fpubh.2022.86923835812486
     [Google Scholar]
  45. Al-rawashdehM. KeikhosrokianiP. BelatonB. AlawidaM. ZwiriA. IoT adoption and application for smart healthcare: A systematic review.Sensors20222214537710.3390/s2214537735891056
     [Google Scholar]
  46. BovenizerW. ChetthamrongchaiP. A comprehensive systematic and bibliometric review of the IoT-based healthcare systems.Cluster Comput.20232653291331710.1007/s10586‑023‑04047‑137359057
     [Google Scholar]
  47. DuH. Deep-learning radiomics based on ultrasound can objectively evaluate thyroid nodules and assist in improving the diagnostic level of ultrasound physicians.Quant. Imaging Med. Surg.20241485932594510.21037/qims‑23‑1597
     [Google Scholar]
  48. SongA-l. Dynamic nomogram for predicting the risk of perioperative neurocognitive disorders in adults.Anesth. Analg.202313761257126910.1213/ANE.0000000000006746
     [Google Scholar]
  49. GuoL. ShiP. ChenL. ChenC. DingW. Pixel and region level information fusion in membership regularized fuzzy clustering for image segmentation.Inf. Fusion20239247949710.1016/j.inffus.2022.12.008
     [Google Scholar]
/content/journals/cmir/10.2174/0115734056380084250720064859
Loading
CT-based 3D Super-resolution Radiomics for the Differential Diagnosis of Brucella vs. Tuberculous Spondylitis using Deep Learning
Curr. Med. Imaging 21, E15734056380084 (2025); https://doi.org/10.2174/0115734056380084250720064859
/content/journals/cmir/10.2174/0115734056380084250720064859
/content/journals/cmir/10.2174/0115734056380084250720064859
Loading

Data & Media loading...

Supplements

Supplementary material is available on the publisher’s website along with the published article.

file format download Download

  • Article Type:     Research Article
Keyword(s): Brucella Spondylitis; Brucellosis; CT imaging; Deep learning radiomics; Super-Resolution; Tuberculosis; Tuberculous Spondylitis

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test