COMPUTER AIDED THERAGNOSIS BASED ON TUMOUR VOLUMETRIC INFORMATION IN BREAST CANCER

Abstract

A computer-assisted technology has recently proposed for the assessment of therapeutic responses to neoadjuvant chemotherapy in patients with locally advanced breast cancer (LABC). The system, however, extracted features from individual scans in a tumour irrespective of its relation to the other scans of the same patient, ignoring the volumetric information. This study addresses this problem by introducing a novel engineered texton-based method in order to account for volumetric information in the design of textural descriptors to represent tumour scans. Methods: A noninvasive computer-aided-theragnosis (CAT) system was developed by employing multi-parametric quantitative ultrasound spectral and backscatter coefficient maps. The proceed was composed of two subdictionaries: one built on the “pre-treatment” and another on “week N” scans, where N was 1, 4, or 8. The learned dictionary of each patient was subsequently used to compute the model (histogram of textons) for each scan of the patient. Advanced machine learning techniques including a kernel-based dissimilarity measure to estimate the distances between “pre-” and “mid-treatment” scans as an indication treatment effectiveness, learning from imbalanced data, and supervised learning were subsequently employed on the texton based features. Results: The performance of the CAT system was tested using statistical tests of significance and leave-one-subject out classification on 56 LABC patients. The proposed textonbased CAT system indicated significant differences in changes between the responding and non-responding patient populations and achieved a high accuracy, sensitivity, and specificity in discriminating between the two patient groups early after the start of treatment, i.e., on weeks 1 and 4 of several months of treatment. Specifically, the CAT system achieved the area under curve (AUC) of 0.81, 0.83 and 0.85 on weeks 1, 4, and 8, respectively. Conclusion: The proposed texton-based CAT system accounted for the volumetric information in “pre-” and “midtreatment” scans of each patient. It was demonstrated that this attribute of the CAT system could boost its performance compared to the cases that the features were extracted from solely individual scans.

Existing System

Maximum mean discrepancy (MMD), diffuse optical imaging (DOI)

Proposed System

1) The design and development of a complete non-invasive computer-aided-theragnosis system based on QUS meth ods, which accounted for the volumetric information in multiple scans of each patient, by using a data-driven texture method. 2) Novel engineering a texton-based method specifically adapted to the application of cancer response monitoring by building one dictionary/codebook per patient, and learning the scan models based on the patient-specific dictionary. 3) Quantifying the different ultrasound characteristics of responding tumours compared to non-responding tumours as early as week 1 in patients with LABC receiving neoadjuvant chemotherapy when using the proposed CAT system.

Architecture

BLOCK DIAGRAM