MSH-5: malignant knee tumors timely detection index
Related Experiment Video For Bone neoplasms
Updated: Aug 2, 2025
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
Gustavo Rivera-Saldívar1, Rubén Torres-González2, Alejandra Cario-Méndez1
1Servicio de Tumores Óseos.
Background:
Bone neoplasms are usually misdiagnosed causing a delay in their treatment. Bone neoplasms are usually confused with tendinitis, 31% of the cases corresponds to osteosarcomas and in 21% to Ewing's sarcomas.
Objective:
To create a clinical-radiographic instrument of high diagnostic suspicion of knee bone neoplasms to prevent a delay in diagnosis.
Method:
A clinimetric study (sensitivity, consistency and validity) was performed in the bone tumor service, Hospital de Ortopedia de la Unidad Médica de Alta Especialidad Dr. Victorio de la Fuente Narváez, Instituto Mexicano del Seguro Social, in México City.
Results:
Characteristics of 153 patients were collected. For the sensitivity phase, 3 domains (signs, symptoms, and radiology) and 12 items were included. Consistency was evaluated with ICC (0.944), 95%CI (0.865-0.977), p < 0.001 and a-Cronbach (0.863). Index obtained a sensitivity of 0.80 and a specificity of 0.882 were obtained. The positive predictive value of the test was 66.6% and the negative predictive value was 93.75%. The positive likelihood ratio was 6.8 and the negative likelihood ratio was 0.2. Validity was evaluated using r-Pearson (0.894; p < 0.001).
Conclusions:
A high suspicion clinical-radiographic index was designed to detect malignant knee tumors with adequate sensitivity, specificity, appearance, content, criteria, and construct validity.
