Breast Tumour Classification Using Ultrasound Elastography with Machine Learning: A Systematic Scoping Review

Ultrasound elastography can quantify stiffness distribution of tissue lesions and complements conventional B-mode ultrasound for breast cancer screening. Recently, the development of computer-aided diagnosis has improved the reliability of the system, whilst the inception of machine learning, such as deep learning, has further extended its power by facilitating automated segmentation and tumour classification. The objective of this review was to summarize application of the machine learning model to ultrasound elastography systems for breast tumour classification. Review databases included PubMed, Web of Science, CINAHL, and EMBASE. Thirteen (n = 13) articles were eligible for review. Shear-wave elastography was investigated in six articles, whereas seven studies focused on strain elastography (5 freehand and 2 Acoustic Radiation Force). Traditional computer vision workflow was common in strain elastography with separated image segmentation, feature extraction, and classifier functions using different algorithm-based methods, neural networks or support vector machines (SVM). Shear-wave elastography often adopts the deep learning model, convolutional neural network (CNN), that integrates functional tasks. All of the reviewed articles achieved sensitivity ³ 80%, while only half of them attained acceptable specificity ³ 95%. Deep learning models did not necessarily perform better than traditional computer vision workflow. Nevertheless, there were inconsistencies and insufficiencies in reporting and calculation, such as the testing dataset, cross-validation, and methods to avoid overfitting. Most of the studies did not report loss or hyperparameters. Future studies may consider using the deep network with an attention layer to locate the targeted object automatically and online training to facilitate efficient re-training for sequential data.

Estimation of Clinical Size of Breast Tumour Lesions Using Contrast Enhanced Magnetic Resonance Imaging: Delineation of Tumour Boundaries

This study covers an extended spectrum of clinical cases (1) to analyze the accurate tumour size, (2) to demarcate accurate tumour boundaries in order to plan an effective target volumes for radiotherapy, thermal ablation including radiofrequency ablation and nanoparticles induced thermal ablation. Once the clinical size of the tumour is established, realistic three-dimensional volume of the tumour can be calculated. Accurate margins (0.5 cm -1 cm) can only be sacrificed if true tumour boundaries with irregular nodal spread can be retrieved.

A Rare Case of Myoid Hamartoma Masquerading As Invasive Breast Carcinoma

Breast Myoid Hamartoma (MH) is a rare type of neoplasm with a poorly understood pathogenesis. Very few literatures have reported such disease with an unclear prognosis and malignant potentiality. Some isolated studies have shown that breast Myoid Hamartoma (MH) may be genetically related to other types of tumours with the involvement of HMGA2 gene. We reported a case of a 64-year-old post-menopausal lady with an underlying chronic idiopathic axonal polyneuropathy (CIAP) that was referred to our centre for a suspected right breast tumour. Clinical and imaging proved the disease to be malignant, however, core biopsy results showed otherwise. Ultrasound of the right breast showed a solid mass with a hypoechoic heterogeneous echotexture and posterior shadowing. A Mammogram highlighted a dense lesion in the right breast with radiolucent halo and macrocalcification. It was reported as BIRADS 4 category. Managing breast Myoid Hamartoma (MH) is proved to be of great challenge to clinicians as meticulous clinical acumen is needed to strategize a proper plan and management, most importantly, not to overlook the disease as the malignant transformation has been reported before.

Diagnostic Value of Vascular Endothelial Growth Factor C (VEGF-C) and CA 15-3 in Breast Cancer

ABSTRACT Background: expression of VEGF-C and CA 15-3 may be useful to differentiate between malignant and benign breast tumour because VEGF-C plays a role in promoting angiogenesis and lymphangiogenesis in malignant processes and CA 15-3 is the soluble form of transmembrane protein MUC1, a tumour marker which shows higher expression in breast cancer.Objective: to determine the diagnostic value of VEGF-C and CA 15-3 as tumour markers in patients with breast cancer.Methods: this diagnostic study recruited 76 patients that underwent surgical biopsy procedures at Dr. Kariadi and Pantiwilasa Citarum Hospitals Semarang. The VEGF-C and CA 15-3 levels in blood specimens taken before surgical biopsy procedure was determined using ELISA method. An ROC curve and AUC were used to establish the cut-off points and diagnostic value. Pathology examination results from the biopsy specimens were used as the gold standard.Results: the cut-off value for VEGF-C and CA 15-3 were 989.50 pg/mL and 74.00 U/mL. Sensitivity for VEGF-C, CA 15-3 and VEGF-C+CA 15-3 were 76.6%, 64.1% and 89.1%. Specificity for VEGF-C, CA 15-3 and VEGF-C+CA 15-3 were 75.0%, 75.0% and 50.0%. The AUC for VEGF-C, CA 15-3 and VEGF-C+CA 15-3 was 0.831 (95% CI = 0.727-0.934), 0.742 (95% CI = 0.628-0.856) and 0.840 (95% CI = 0.742-0.938).Conclusion: VEGF-C in combination with CA 15-3 is the best diagnostic parameter for breast cancer and has the best accuracy as a tumour marker for breast cancer.

The Profile of Expression of SG2NAs Differs Between Cancer Types and it is Involved in the Reprogramming of Tumour Cell Proteome.

Striatin and SG2NA are scaffold proteins that from signalling complexes called STRIPAK. It has been associated with cancer and other diseases. Our earlier studies have shown that SG2NA forms a complex with the cancer associate protein DJ-1 and signalling kinase Akt, promoting cancer cell survival. In the present study, we used bioinformatics analyses to confirm the existence of two isoforms of human SG2NA i.e., 78 and 87 kDas. In addition, several smaller isoforms like 35 kDa were also seen in western blot analyses of human cell lysates. The expression of these isoforms varies between different human cancer cell lines. Also, the protein level does not corroborate with its transcript level, suggesting a complex regulation of its expression. In breast tumour tissues, the expression of the 35 and 78 kDa isoforms was higher as compared to the adjacent normal tissues, while the 87 kDa isoform was detected in the breast tumour tissues only. With the progression of stages of breast cancer, the expression of 78 kDa isoform decreased, while 87 kDa became undetectable. In coimmunoprecipitation assay, the profile of SG2NA interactome in breast tumor vis-à-vis adjacent normal breast tissues shows hundreds of common proteins, while some proteins specifically interacted in breast tumour tissue only. We conclude that SG2NA is involve in diverse cellular pathways and have roles in cellular reprogramming during tumorigenesis.

Repeat Breast-Conserving Surgery Versus Salvage Mastectomy for Ipsilateral Breast Tumour Recurrence After Breast-Conserving Surgery in Breast Cancer Patients: A Meta-Analysis

BackgroundSalvage mastectomy (SM) is the standard surgery for ipsilateral breast tumour recurrence (IBTR). However, whether repeat breast-conserving surgery (RBCS) is an alternative method remains unclear. We performed a meta-analysis to compare the effects of RBCS and SM after IBTR for breast-conserving surgery (BCS).MethodsWe searched PubMed, Cochrane, Wiley Online and Embase for controlled studies comparing RBCS and SM after IBTR for BCS (published between 1993 and 2019, published in English). Our main endpoints were the secondary local recurrence rate (SLRR), distant metastasis rate (DMR) and overall survival (OS). We used a random-effects model or fixed-effects model for data pooling.ResultsFifteen of the 424 eligible studies were ultimately included, and all studies were retrospective cohort studies (n=2532 participants). 1) SLRR: The SLRR of RBCS was higher than SM (pooled relative rate (pRR) = 1.87, 95% CI 1.22 - 2.86, P=0.004). Stratified analysis was performed according to whether radiotherapy was performed after salvage surgery (radiotherapy group: 2ndRT, no radiotherapy group: no-2ndRT), and the following results were revealed: pRR=0.43 (95% CI 0.20-0.95, P=0.04) for group 2ndRT; and pRR=2.30 (95% CI 1.72-3.06, P<0.00001) for group no-2ndRT. These results showed that the main cause of heterogeneity was salvage radiotherapy. 2) DMR: No significant difference in the DMR was observed between RBCS and SM (pRR = 0.61, 95% CI 0.37 - 1.01, P=0.05). 3) OS: No significant difference in OS was observed between RBCS and SM (pRR=0.65, 95% CI 0.39 - 1.08, P=0.10).ConclusionsThe SLRR of RBCS was higher than SM for ITBR after BCS, but survival was not affected. RBCS may be used as an alternative for IBTR patients after BCS with strict control for several indications, such as tumor size, recurrence interval and biological behavior, and attaching importance to subsequent salvage radiotherapy and systematic therapy.