Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · physics/0701053

An Automatic System to Discriminate Malignant from Benign Massive Lesions on Mammograms

A. Retico·P. Delogu·M. E. Fantacci·P. Kasae
View original source ↗Read source document ↗Export metadata ↓

Abstract

Mammography is widely recognized as the most reliable technique for early detection of breast cancers. Automated or semi-automated computerized classification schemes can be very useful in assisting radiologists with a second opinion about the visual diagnosis of breast lesions, thus leading to a reduction in the number of unnecessary biopsies. We present a computer-aided diagnosis (CADi) system for the characterization of massive lesions in mammograms, whose aim is to distinguish malignant from benign masses. The CADi system we realized is based on a three-stage algorithm: a) a segmentation technique extracts the contours of the massive lesion from the image; b) sixteen features based on size and shape of the lesion are computed; c) a neural classifier merges the features into an estimated likelihood of malignancy. A dataset of 226 massive lesions (109 malignant and 117 benign) has been used in this study. The system performances have been evaluated terms of the receiver-operating characteristic (ROC) analysis, obtaining A_z = 0.80+-0.04 as the estimated area under the ROC curve.

Explore related subjects

physics.med-ph

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

A. Retico, P. Delogu, M. E. Fantacci, P. Kasae. 2007-01-04. An Automatic System to Discriminate Malignant from Benign Massive Lesions on Mammograms. https://doi.org/10.1016/j.nima.2006.08.093

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Light-sheet microscopy in cancer pathology: current applications, clinical challenges and future perspectives

Light-sheet fluorescence microscopy (LSFM) has emerged as a powerful research technology for three-dimensional (3D) visualization of biological tissues and is increasingly applied to human cancer specimens, including optically cleared biopsies and surgical resection samples. In this Perspective article, we review current applications of LSFM in oncology, and discuss how recent developments in microscopy, tissue preparation, labelling, and computational analysis are expanding the possibilities of 3D cancer pathology. Although LSFM has demonstrated the feasibility of large-volume imaging in selected human specimens, its application remains predominantly within research and translational settings and is not currently part of routine clinical pathology. We therefore distinguish demonstrated applications from potential future uses. We discuss how LSFM could eventually contribute to selected diagnostic and surgical workflows by providing 3D information on tumour architecture, spatial heterogeneity, metastatic dissemination, and surgical margins. These applications remain prospective and will require substantial technical, clinical, regulatory, computational, and workflow validation before clinical implementation can be considered. Ultimately, LSFM may complement rather than replace conventional histopathology by extending pathological assessment from two-dimensional tissue sections towards 3D spatial analysis.

physics.med-ph↗
arXiv2026

Quantitative Ultrasound Perfusion Biomarker: Cross-Modality Validation with Laser Speckle Imaging

Rationale and Objectives: Tooth extraction can cause alveolar bone resorption and variable healing despite socket preservation. Current perfusion assessment relies largely on visual evaluation, limiting objective monitoring of socket healing. We developed and validated a quantitative color flow ultrasound framework to reduce sensitivity to imaging settings and motion artifacts. Materials and Methods: Color flow ultrasound data were acquired longitudinally in 30 patients undergoing tooth extraction and grafting. Velocity- and power-weighted color pixel density (CPDvel and CPDpwr) were derived from complex-valued color flow (TRI) data at 0-1.5 mm beneath the epithelial surface. Gain robustness was assessed using a Quantitative Imaging Biomarker Alliance (QIBA) flow phantom. Concurrent laser speckle contrast imaging (LASCA) provided in vivo comparison. Results: Phantom velocity estimates remained stable across color gain settings of 20-100 a.u. at flow rates of 2, 4, 6, and 8 mL/s. Perfusion metrics decreased over approximately 2-30 weeks after extraction. Nonlinear fits yielded R2 values of 0.74 for CPDvel, 0.45 for CPDpwr, and 0.66 for LASCA. Exponential half-decay times were 60.3, 92.4, and 48.3 days, respectively, relative to the first post-extraction measurement. LASCA correlated linearly with ultrasound-derived velocity (R2 = 0.68) and power (R2 = 0.74; both p < 0.05). Conclusions: The framework supports noninvasive, quantitative monitoring of superficial perfusion during socket healing with reduced sensitivity to color gain. Direct analysis of TRI data provides depth-specific perfusion metrics and may improve consistency across imaging sessions. Further validation is needed to determine its ability to identify compromised healing and inform clinical decisions.

physics.med-ph↗
arXiv2026

LesionStudio: A Patient-Informed Platform for Task-Specific Optimization of Emission Tomography Reconstructions

Purpose: For task-specific PET and SPECT reconstruction optimization, we developed LesionStudio, an open-source platform incorporating patient background heterogeneity, realistic lesion morphology and texture, and lesion-background interactions beyond conventional phantoms. Methods: LesionStudio, a Python framework built on PyTomography, inserts lesions into PET/SPECT data using modality-specific forward projection and stochastic background ablation. Clinical realism was evaluated by physician review of 30 inserted lesions in eight DLBCL patients from a 1,579-lesion FDG lymphoma library. Quantification and detectability were assessed in FDG PET using spherical and patient-derived lesions varying in size, LBR, and site, reconstructed with OSEM and BSREM. Results: Inserted lesions were visually indistinguishable from true disease (AUC = 0.49). OSEM improved the patient-derived quantification limit from ~3.7 mL at iteration 4 to ~2.0 mL at iteration 10, but with increased noise and edge artifacts. Weakly regularized BSREM (beta = 5) achieved a similar ~2.0 mL limit with stability, whereas stronger regularization increased under-recovery. Detectability was contrast-dependent: no lesions were detected at LBR 2:1, whereas detectable volume ranged from 0.70 to 3.63 mL at LBR 4:1 and decreased to 0.70 mL at LBR 6:1-8:1 for both OSEM and weakly regularized BSREM. Conclusion: Reconstruction settings can shift SUV metrics and detectability enough to affect interpretation, so no single protocol suits every task. OSEM required early stopping for detection and feature stability but more iterations for mean-SUV quantification, whereas weakly regularized BSREM matched both at one setting. LesionStudio (https://github.com/qurit/LesionStudio), with our public library of realistic lesions, provides a reproducible framework for task-specific optimization across radiopharmaceuticals and scanners.

physics.med-ph↗