METHODS: Breast cancer MRI images were classified into BA, BF, BPT, BTA, MDC, MLC, MMC, and MPC using a proposed Deep Learning model with additional 5 fine-tuned Deep learning models consisting of Xception, InceptionV3, VGG16, MobileNet and ResNet50 trained on ImageNet database. The dataset was collected from Kaggle depository for breast cancer detection and classification. That Dataset was boosted using GAN technique. The images in the dataset have 4 magnifications (40X, 100X, 200X, 400X, and Complete Dataset). Thus we evaluated the proposed Deep Learning model and 5 pre-trained models using each dataset individually. That means we carried out a total of 30 experiments. The measurement that was used in the evaluation of all models includes: F1-score, recall, precision, accuracy.
RESULTS: The classification F1-score accuracies of Xception, InceptionV3, ResNet50, VGG16, MobileNet, and Proposed Model (BCCNN) were 97.54%, 95.33%, 98.14%, 97.67%, 93.98%, and 98.28%, respectively.
CONCLUSION: Dataset Boosting, preprocessing and balancing played a good role in enhancing the detection and classification of breast cancer of the proposed model (BCCNN) and the fine-tuned pre-trained models' accuracies greatly. The best accuracies were attained when the 400X magnification of the MRI images due to their high images resolution.
OBJECTIVE: This study aimed to determine the survival rate of breast cancer among the women of Malaysia and characteristics of the survivors.
METHOD: A retrospective cohort study was conducted on secondary data obtained from the Breast Cancer Registry and medical records of breast cancer patients admitted to Hospital Kuala Lumpur from 2005 to 2009. Survival data were validated with National Birth and Death Registry. Statistical analysis applied logistic regression, the Cox proportional hazard model, the Kaplan-Meier method and log rank test.
RESULTS: A total of 868 women were diagnosed with breast cancer between January 2005 and December 2009, comprising 58%, 25% and 17% Malays, Chinese and Indians, respectively. The overall survival rate was 43.5% (CI 0.573-0.597), with Chinese, Indians and Malays having 5 year survival rates of 48.2% (CI 0.444-0.520), 47.2% (CI 0.432-0.512) and 39.7% (CI 0.373-0.421), respectively (p<0.05). The survival rate was lower as the stages increased, with the late stages were mostly seen among the Malays (46%), followed by Chinese (36%) and Indians (34%). Size of tumor>3.0cm; lymph node involvement, ERPR, and HER 2 status, delayed presentation and involvement of both breasts were among other factors that were associated with poor survival.
CONCLUSIONS: The overall survival rate of Malaysian women with breast cancer was lower than the western figures with Malays having the lowest because they presented at late stage, after a long duration of symptoms, had larger tumor size, and had more lymph nodes affected. There is an urgent need to conduct studies on why there is delay in diagnosis and treatment of breast cancer women in Malaysia.
PRESENTATION OF CASE: We present the case of a female patient with a history of invasive lobular breast carcinoma who had undergone mastectomy and axillary dissection, followed by chemoradiotherapy. Over the next nine years, she developed ovarian and bone metastases for which appropriate treatment was provided. A right iliac fossa mass was discovered during routine clinic review, though she remained asymptomatic. Computed tomography scan showed ileocecal intussusception. Histopathological examination of the right hemicolectomy specimen following emergency surgery confirmed metastatic invasive lobular carcinoma to the GI tract.
DISCUSSION: GI tract metastasis may present 30 years after the primary breast cancer. Up to 20% of patients may be asymptomatic as shown by Montagna et al. When present, symptoms are commonly non-specific and vague. Histological diagnosis is challenging. GI metastasis typically appears as intramural infiltration of the bowel wall by small cells arranged in cords.
CONCLUSION: It is important to maintain a suspicion for GI tract metastasis in breast cancer patients who present with abdominal mass or GI symptoms, as this aids in prompt institution of accurate and appropriate management.
MATERIALS AND METHODS: Twenty-four patients with clinically node-negative breast cancer were recruited. Combined radiotracer and blue dye methods were used for identification of SLNs. The nodes were thinly sliced and embedded. Serial sectioning and immunohistochemical (IHC) staining against AE1/AE3 were performed if initial HandE sections of the blocks were negative.
RESULTS: SLNs were successfully identified in all patients. Ten cases had nodal metastases with 7 detected in SLNs and 3 detected only in axillary nodes (false negative rate, FNR=30%). Some 5 out of 7 metastatic lesions in the SLNs (71.4%) were detected in initial sections of the thinly sliced tissue. Serial sectioning detected the remaining two cases with either micrometastases or isolated tumour cells (ITC).
CONCLUSIONS: Thin slicing of tissue to 3-5mm thickness and serial sectioning improved the detection of micro and macro-metastases but the additional burden of serial sectioning gave low yield of micrometastases or ITC and may not be cost effective. IHC validation did not further increase sensitivity of detection. Therefore its use should only be limited to confirmation of suspicious lesions. False negative cases where SLNs were not involved could be due to skipped metastases to non-sentinel nodes or poor technique during procurement, resulting in missed detection of actual SLNs.
METHODS: Patients diagnosed with invasive breast cancer (BC) from 2005 to 2013 at our tertiary institution were included and divided according to race and subtypes. Demographic and clinical information of non-metastatic TNBC patients were analyzed. Log-rank test, univariate and multivariate Cox proportional hazard regression models were used to find associated risk factors related with overall survival (OS) and disease-free survival (DFS).
RESULTS: Among 1227 BC patients, 129 (10.5%) had TNBC. TNBC patients had the worst OS (P: 0.0005) and DFS (P: 0.0016) among the subtypes. However, variations in race did not have any difference in OS or DFS among TNBC patients. Axillary lymph node involvement, invasive lobular histology, larger tumor size, and the presence of lymphovascular invasion (LVI) were factors associated with both poor DFS and OS among TNBC patients.
CONCLUSIONS: Racial variation did not have any impact on the prognosis of the TNBC.
MATERIALS AND METHODS: Patients who were treated with first line palliative chemotherapy for de novo MBC from 2002-2011 in UMMC were identified from the UMMC Breast Cancer Registry. Information collected included patient demographics, histopathological features, treatment received, including the different chemotherapy regimens, and presence of FN and TRD. FN was defined as an oral temperature >38.5° or two consecutive readings of >38.0° for 2 hours and an absolute neutrophil count <0.5x109/L, or expected to fall below 0.5x109/L (de Naurois et al, 2010). TRD was defined as death occurring during or within 30 days of the last chemotherapy treatment, as a consequence of the chemotherapy treatment. Statistical analysis was performed using the SPSS version 18.0 software. Survival probabilities were estimated using the Kaplan-Meier method and differences in survival compared using log-rank test.
RESULTS: Between 1st January 2002 and 31st December 2011, 424 patients with MBC were treated in UMMC. A total of 186 out of 221 patients with de novo MBC who received first line palliative chemotherapy were analyzed. The mean age of patients in this study was 49.5 years (range 24 to 74 years). Biologically, ER status was negative in 54.4% of patients and Her-2 status was positive in 31.1%. A 5-flourouracil, epirubicin and cyclophosphamide (FEC) chemotherapy regimen was chosen for 86.6% of the cases. Most patients had multiple metastatic sites (58.6%). The main result of this study showed a FN rate of 5.9% and TRD rate of 3.2%. The median survival (MS) for the entire cohort was 19 months. For those with multiple metastatic sites, liver only, lung only, bone only and brain only metastatic sites, the MS was 18, 24, 19, 24 and 8 months respectively (p-value= 0.319).
CONCLUSIONS: In conclusion, we surmise that FEC is a safe regimen with acceptable FN and TRD rates for de novo MBC.