There are millions of cancer cases worldwide every year, and breast cancer is one of the most prevalent diseases with the highest mortality rate. The manual effort of pathologists can be significantly reduced by computerized diagnostic systems, which improve the accuracy and reliability of diagnosis. In this paper, we present four novel systems for breast cancer diagnosis in four different scenarios: binary versus multi-class classification and magnification-specific (MS) versus magnification-independent (MI) classification. In each of the proposed systems, we developed an automatic score-level fused CNN model using a pretrained deep neural network and named it the Multi-Level Feature Fusion (MLF2) model. The MLF2-CNN, similar to the conventional CNN models, integrates the feature extraction and classification phases of BC classification into a single automatic learning procedure. Additionally, MLF2-CNN performs an automatic score-level fusion of several classifiers that were trained with multi-level features to make the final decision. A pretrained DenseNet-121 is selected as the backbone of the proposed MLF2-CNN, and several new links are added to the CNN architecture to capture multi-stage features. Several experiments on the publicly available BreakHis dataset demonstrate that the proposed systems capture the best descriptive features and outperform state-of-the-art techniques in most of the scenarios.

Biomarkers that predict disease progression and treatment-free remission (TFR) would be of significant clinical value in chronic myeloid leukaemia (CML). We have previously shown that CIP2A levels at diagnosis can identify patients at increased risk of progression. One mechanism by which CIP2A acts is through upregulation of the anti-apoptotic gene BCL-XL. In this study, we evaluated BCL-XL mRNA expression as a diagnostic biomarker using samples from the SPIRIT2 and DESTINY clinical trials. In SPIRIT2, which compared imatinib and dasatinib as first-line therapies, high BCL-XL expression was associated with treatment failure, poor early molecular response, and lower rates of MR2 and MR3 achievement in patients treated with imatinib. In the DESTINY trial, which assessed treatment de-escalation and discontinuation, BCL-XL expression was significantly higher in patients who experienced molecular relapse compared to those achieving sustained TFR. Notably, increases in BCL-XL were detectable 6 to 8 months prior to molecular relapse, suggesting it may serve as an early biomarker of unsuccessful TFR. We now propose a clinical diagnostic toolkit combining CIP2A and BCL-XL biomarkers to stratify CML patients by the risk of disease progression and likelihood of achieving successful TFR.