A Computer-aided Framework for Detecting Osteosarcoma in Computed Tomography Scans

Maximo Rodriguez-Herrero; Dante D. Sanchez-Gallegos; Marco Antonio Núñez-Gaona; Heriberto Aguirre-Meneses; Luis Alberto Villalvazo Gutiérrez; Mario Ibrahin Gutiérrez Velasco; J. L. Gonzalez-Compean; Jesus Carretero

cs.CVMar 10, 2026

A Computer-aided Framework for Detecting Osteosarcoma in Computed Tomography Scans

Authors: Maximo Rodriguez-Herrero, Dante D. Sanchez-Gallegos, Marco Antonio Núñez-Gaona, Heriberto Aguirre-Meneses, Luis Alberto Villalvazo Gutiérrez, Mario Ibrahin Gutiérrez Velasco, J. L. Gonzalez-Compean, Jesus Carretero

Organizations: Universidad Carlos III de Madrid, Departamento de Informática, ARCOS, Avenida de la Universidad, 30 (edificio Sabatini), 28911 Leganés (Madrid), España · Instituto Nacional de Rehabilitacion “Luis Guillermo Ibarra Ibarra”, Mexico City, México · SECIHTI - Instituto Nacional de Rehabilitacion “Luis Guillermo Ibarra Ibarra”, Mexico City, México · Cinvestav Tamaulipas, Cd. Victoria, Tamaulipas, México

Abstract

Osteosarcoma is the most common primary bone cancer, mainly affecting the youngest and oldest populations. Its detection at early stages is crucial to reduce the probability of developing bone metastasis. In this context, accurate and fast diagnosis is essential to help physicians during the prognosis process. The research goal is to automate the diagnosis of osteosarcoma through a pipeline that includes the preprocessing, detection, postprocessing, and visualization of computed tomography (CT) scans. Thus, this paper presents a machine learning and visualization framework for classifying CT scans using different convolutional neural network (CNN) models. Preprocessing includes data augmentation and identification of the region of interest in scans. Post-processing includes data visualization to render a 3D bone model that highlights the affected area. An evaluation on 12 patients revealed the effectiveness of our framework, obtaining an area under the curve (AUC) of 94.8% and a specificity of 94.6%.

Figures & tables

Figure 1: Proposed framework for identifying osteosarcoma using CT images.

Figure 2: Automated pre-processing pipeline applied to each patient’s CT scan. From left to right: raw CT image with scanning table artifacts, result of morphological opening using a disk-shaped structuring element (radius = 10px), binary mask obtained via Otsu thresholding, and final selection of the two largest connected components representing the leg region.

Figure 3: Slice-wise bone extraction pipeline applied to pre-processed CT images. From left to right: Gaussian smoothing ( $\sigma=2$ ), K-means clustering with $k=5$ , selection of the brightest cluster to isolate bone tissue, and final binary mask after morphological closing and hole filling outlined on the pre-processed CT scan.

Panel of Figure 4 — Figure 4: 3D model output of the framework for an osteosarcoma-positive patient, shown from two different viewing angles. The red bounding box delineates the predicted affected region.

Explore similar work

Integrating Local Detail and Global Context: A Dual-Input Multi-Task Learning Framework for Bone Tumor Diagnosis
Sep 23, 2026S. M. Nasif Uddin, Rusab Sarmun, Muhammad E. H. Chowdhury +3Cross-Attention Multi-Task Learning
Department of Electrical and Electronic Engineering, Ahsanullah University of Science and Technology, Bangladesh · Department of Electrical and Electronic Engineering, University of Dhaka, Bangladesh · Department of Electrical Engineering, Qatar University, Doha 2713, Doha, Qatar +1
Cross-Stage Attention Multi-Expert Network for Radiologist-Inspired Breast Ultrasound Diagnosis
May 25, 2026Xinyang Zhai, Chong Yang, Ruizhi ZhangMixture of Experts Medical Image Classification
Bowel Obstruction Detection and Localization on Abdominal CT with Deep Learning
Jul 24, 2026Moritz Vandenhirtz, Andrea Agostini, Dana Belde +5Medical Image Anomaly Detection Explainable Medical Image Analysis
Department of Computer Science, ETH Zurich, Zurich, Switzerland · Department of Radiology, Kantonsspital Baden, affiliated Hospital for Research and Teaching of the Faculty of Medicine of the University of Zurich, Baden, Switzerland · Department of Forensic Medicine Zurich, University of Zurich, Zurich, Switzerland

Sep 23, 2026J/K move · Enter open · S save

Integrating Local Detail and Global Context: A Dual-Input Multi-Task Learning Framework for Bone Tumor Diagnosis

S. M. Nasif Uddin, Rusab Sarmun, Muhammad E. H. Chowdhury, Adam Mushtak+2

Department of Electrical and Electronic Engineering, Ahsanullah University of Science and Technology, Bangladesh · Department of Electrical and Electronic Engineering, University of Dhaka, Bangladesh · Department of Electrical Engineering, Qatar University, Doha 2713, Doha, Qatar+1

Cross-Attention Multi-Task Learning Medical Image Classification Medical Image Segmentation

Open paper AskPDF arXiv

Primary bone tumors are rare but clinically aggressive neoplasms whose diagnosis from radiographs is challenged by heterogeneous morphology, subtle lesion margins, and overlapping bone structures. To address the limitations of existing single-view models, we present a dual-input, multi-task learning framework that, to our knowledge, is the first to apply bidirectional cross-modal attention between a lesion crop and the full radiograph for joint segmentation and subtype classification. Using the multi-institutional Bone Tumor X-ray Radiograph Dataset (BTXRD, n=3,746), we employ a YOLO-based detector to generate regions of interest, which are paired with full images as inputs to a dual-stream DenseNet121 architecture. Features are integrated via a novel cross-modal attention fusion strategy, refined by Hierarchical Multi-scale Feature Fusion, effectively balancing fine-grained lesion detail with global anatomical context. Evaluated on a held-out patient-level test split, the model demonstrates superior performance over single-input baselines, achieving an overall Dice Similarity Coefficient of 0.896 and a macro-averaged classification F1-score of 0.928. Notably, the system exhibits exceptional sensitivity for malignant osteosarcoma (AUC 0.999), validating the potential of dual-stream context modeling to support radiologists in accurate, early decision-making.

Abstract

Explore similar work

Integrating Local Detail and Global Context: A Dual-Input Multi-Task Learning Framework for Bone Tumor Diagnosis

Cross-Stage Attention Multi-Expert Network for Radiologist-Inspired Breast Ultrasound Diagnosis

Bowel Obstruction Detection and Localization on Abdominal CT with Deep Learning