Improved mammographic interpretation of masses using computer-aided diagnosis

¹Department of Electro-Optics, Jerusalem College of Technology, P. O. Box 16031, IS-91160 Jerusalem, Israel.

|February 9, 2000

Summary

Computer-aided diagnosis (CAD) significantly improves mammogram interpretation accuracy. This system enhances image visualization and uses quantitative analysis to help radiologists differentiate benign from malignant mass lesions.

Area of Science:

Radiology
Medical Imaging
Artificial Intelligence in Medicine

Background:

Mammography is crucial for breast cancer detection.
Interpreting mammographic mass lesions can be challenging, impacting diagnostic accuracy.
Computer-aided diagnosis (CAD) systems offer potential improvements in image analysis.

Purpose of the Study:

To evaluate computerized image enhancement in mammography.
To identify criteria for distinguishing benign from malignant findings using CAD.
To assess the role of quantitative analysis in improving mass lesion interpretation.

Main Methods:

Digitized mammographic mass lesions were analyzed using a prototype CAD system with image enhancement.
Quantitative features, such as spiculation, were automatically extracted.
A pattern recognition scheme was trained on retrospective data and applied to prospective cases.
Receiver operating characteristics (ROC) curves were used to analyze diagnostic accuracy.

Main Results:

CAD system significantly improved specificity (14% to 50%) and positive predictive value (0.47 to 0.62).
The area under the ROC curve (A(z)) increased from 0.66 (conventional) to 0.81 (CAD-assisted).
The pattern recognition scheme achieved a higher A(z) of 0.95.

Conclusions:

Computer-aided diagnosis (CAD) enhances diagnostic accuracy in mammography.
Objective quantitative features from digitized mammograms aid in differentiating benign and malignant masses.
CAD systems can effectively assist radiologists in interpreting mammographic mass lesions.

Related Experiment Videos

Improved mammographic interpretation of masses using computer-aided diagnosis

Related Concept Videos

The influence of mouth breathing on skeletal and dental features of splanchnocranium.

Do we have to reduce the recall period? Validity of a daily physical activity questionnaire (PAQ24) in young active adults.

Somatic mutations in kinetochore gene KNSTRN are associated with basal proliferating actinic keratoses and cutaneous squamous cell carcinoma.

Enhancing social functioning in young people at Ultra High Risk (UHR) for psychosis: A pilot study of a novel strengths and mindfulness-based online social therapy.

The influence of different illumination parameters on protoporphyrin IX induced cell death in squamous cell carcinoma cells.

Patient-reported health outcomes in patients with non-melanoma skin cancer and actinic keratosis: results from a large-scale observational study analysing effects of diagnoses and disease progression.

Multicentre performance and consistency of two deep learning models for malignancy probability estimation of incidental pulmonary nodules.

Malignancy risk and contrast-enhanced ultrasound guidance in non-diagnostic biopsies of peripheral lung lesions: a multicenter study.

Pericoronary fat radiomics on coronary CT angiography for predicting major adverse cardiac events: a systematic review and meta-analysis.

Region-, zone-, and layer-specific knee cartilage T2 relaxation patterns in sports-active adolescents.

Normal appearance of the parathyroid gland on CT.

Coronary plaque quantification on energy-integrating and photon-counting detector CT: reproducibility and power modeling for mixed-platform trials.

Seeing beyond positivity: computer vision approach to decoding BinaxNOW COVID-19 tests and forecasting negative results.

A Deep Learning Pipeline for Cell Segmentation and Viability Quantification in 3D Constructs From Fluorescence Microscopy Images.

Weighted multi-scale and wavelet-enhanced Segment Anything Model for salient object detection.

Comparison of cone beam computed tomography multiplanar reconstruction in the plane of stapes and the lateral semicircular canal for the diagnosis of otosclerosis.

A Multimodal Dataset for Clinical Gait Analysis compromising Optical Motion Capture and Textual Gait Analyses from Patients undergoing Total Knee Arthroplasty.

Deep Learning-based Automated Vessel Extraction for VR Image Creation From Cerebral TOF-MRA.

Related Experiment Videos

Improved mammographic interpretation of masses using computer-aided diagnosis

Area of Science:

Background:

Purpose of the Study:

Main Methods:

Main Results:

Conclusions:

Related Concept Videos

Related Articles