Scholarlink

<p>Soft tissue sarcomas occur in tissues that support and connect body structures. On Magnetic Resonance Imaging (MRI) scans show variability in appearance because of their high diversity and low-frequency phenomena in the body. Patients cannot get the required care because they are frequently misdiagnosed as other disorders. Researchers have created numerous machine learning models to categorize tumors but must address incorrect diagnoses sufficiently. Here, the suggested model is compared with prevailing models for evaluating such tumors numerically to consider the heterogeneity. As a result, a learning strategy is proposed that combines the feature learning method for pre-processing and classification to eliminate bias and instability deviation and perform classification with the Functional Support Vector Machine (SVM) algorithm over the conventional network model (<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13198_2025_2955_Article_IEq1.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="75" /> </InlineMediaObject> <EquationSource Format="TEX">\(f-SCN\)</EquationSource> </InlineEquation>). Tests on available data demonstrate a considerable advancement over earlier findings with 95% accuracy. This supports the idea that tools for assisting automatic decision-making processes in STT diagnosis could be developed using machine learning approaches.</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 generalization model for soft tissue tumor prediction using a learning approach

  • 期刊论文

  • 发表日期：

    2025年9月12日

查看全文
  • Chandra Sekhar Koppireddy,
  • G. Siva Nageswara Rao

摘要

Soft tissue sarcomas occur in tissues that support and connect body structures. On Magnetic Resonance Imaging (MRI) scans show variability in appearance because of their high diversity and low-frequency phenomena in the body. Patients cannot get the required care because they are frequently misdiagnosed as other disorders. Researchers have created numerous machine learning models to categorize tumors but must address incorrect diagnoses sufficiently. Here, the suggested model is compared with prevailing models for evaluating such tumors numerically to consider the heterogeneity. As a result, a learning strategy is proposed that combines the feature learning method for pre-processing and classification to eliminate bias and instability deviation and perform classification with the Functional Support Vector Machine (SVM) algorithm over the conventional network model ( \(f-SCN\) ). Tests on available data demonstrate a considerable advancement over earlier findings with 95% accuracy. This supports the idea that tools for assisting automatic decision-making processes in STT diagnosis could be developed using machine learning approaches.