Scholarlink

Background <p>Chronic osteomyelitis poses major challenges in orthopaedic practice due to its complex and unpredictable disease progression. Effective treatment requires complete removal of necrotic and infected bone tissue. Surgeons often face two opposing risks: inadequate resection, potentially leading to infection recurrence, and excessive resection, which can cause irreversible bone damage and functional loss. Precise intraoperative localisation of lesions is therefore crucial; however, accurate localisation and quantification of intralesional heterogeneity remain difficult to achieve with conventional imaging techniques.</p> Methods <p>To address these limitations, we developed a novel approach that integrates habitat imaging and radiomics to improve diagnostic accuracy and guide surgical management for chronic osteomyelitis. Using magnetic resonance imaging, we segmented lesions into sub-regions (habitats) using K-means clustering, enabling detailed assessment of various lesion microenvironments. We extracted radiomic features, including geometric, intensity and texture attributes, from each habitat and incorporated them into a diagnostic model using advanced feature selection techniques.</p> Results <p>The model demonstrated strong diagnostic performance, with an area under the curve of 0.983 in the training set and 0.894 in the validation set. It identified potential osteomyelitis lesions that were not visually apparent, providing precise guidance for surgical debridement while minimising healthy bone destruction. Radiomic features related to lesion shape and texture were particularly important for imaging-based diagnosis of osteomyelitis.</p> Conclusions <p>The findings highlight the diagnostic value of integrating habitat imaging and radiomics in chronic osteomyelitis, significantly enhancing diagnostic accuracy and supporting personalised surgical interventions. This approach advances precision medicine and may be applicable to other complex diseases requiring detailed lesion analysis and individualised treatment strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A multicenter study: habitat imaging and radiomics to guide precision and individualized surgical treatment in chronic osteomyelitis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月27日

查看全文
  • Qiyu Jia,
  • Junna Wang,
  • Jie Lin,
  • Rongyu Zhang,
  • Abudusalamu Alimujiang,
  • Qiankun Jin,
  • Jian Guo,
  • Xi Wang,
  • Jun Zhang,
  • Ziyi Xu,
  • Meng Zhao,
  • Zengru Xie,
  • Chuang Ma,
  • Hao Zheng

摘要

Background

Chronic osteomyelitis poses major challenges in orthopaedic practice due to its complex and unpredictable disease progression. Effective treatment requires complete removal of necrotic and infected bone tissue. Surgeons often face two opposing risks: inadequate resection, potentially leading to infection recurrence, and excessive resection, which can cause irreversible bone damage and functional loss. Precise intraoperative localisation of lesions is therefore crucial; however, accurate localisation and quantification of intralesional heterogeneity remain difficult to achieve with conventional imaging techniques.

Methods

To address these limitations, we developed a novel approach that integrates habitat imaging and radiomics to improve diagnostic accuracy and guide surgical management for chronic osteomyelitis. Using magnetic resonance imaging, we segmented lesions into sub-regions (habitats) using K-means clustering, enabling detailed assessment of various lesion microenvironments. We extracted radiomic features, including geometric, intensity and texture attributes, from each habitat and incorporated them into a diagnostic model using advanced feature selection techniques.

Results

The model demonstrated strong diagnostic performance, with an area under the curve of 0.983 in the training set and 0.894 in the validation set. It identified potential osteomyelitis lesions that were not visually apparent, providing precise guidance for surgical debridement while minimising healthy bone destruction. Radiomic features related to lesion shape and texture were particularly important for imaging-based diagnosis of osteomyelitis.

Conclusions

The findings highlight the diagnostic value of integrating habitat imaging and radiomics in chronic osteomyelitis, significantly enhancing diagnostic accuracy and supporting personalised surgical interventions. This approach advances precision medicine and may be applicable to other complex diseases requiring detailed lesion analysis and individualised treatment strategies.