Nonnegative matrix factorization (NMF) is primarily a linear dimensionality reduction technique that factorizes a nonnegative data matrix into two smaller nonnegative matrices: one that represents the basis of the new subspace and the second that holds the coefficients of all the data points in that new space. In principle, the nonnegativity constraint forces the representation to be sparse and parts based. Instead of extracting holistic features from the data, real parts are extracted that should be significantly easier to interpret and analyze. The size of the new subspace selects how many features will be extracted from the data. An effective choice should minimize the noise while extracting the key features. We propose a mechanism for selecting the subspace size by using a minimum description length technique. We demonstrate that our technique provides plausible estimates for real data as well as accurately predicting the known size of synthetic data. We provide an implementation of our code in a Matlab format.
Breast cancer is the leading cause of cancer-related death in women worldwide 1 . Here, in the Breast Cancer-Anti-Progestin Prevention Study 1 (BC-APPS1; NCT02408770 ), we assessed whether progesterone receptor antagonism with ulipristal acetate for 12 weeks reduces surrogate markers of breast cancer risk in 24 premenopausal women. We used multilayered OMICs and live-cell approaches as readouts for molecular features alongside clinical imaging and tissue micromechanics correlates. Ulipristal acetate reduced epithelial proliferation (Ki67) and the proportion, proliferation and colony formation capacity of luminal progenitor cells, the putative cell of origin of aggressive breast cancers 2 . MRI scans showed reduction in fibroglandular volume with treatment, whereas single-cell RNA sequencing, proteomics, histology and atomic force microscopy identified extracellular matrix remodelling with reduced collagen organization and tissue stiffness. Collagen VI was the most significantly downregulated protein after ulipristal acetate treatment, and we uncovered an unanticipated spatial association between collagen VI and SOX9 high luminal progenitor cell localization, establishing a link between collagen organization and luminal progenitor activity. Culture of primary human breast epithelial cells in a stiff environment increased luminal progenitor activity, which was antagonized by anti-progestin therapy, strengthening this mechanistic link. This study offers a template for biologically informed early-phase therapeutic cancer prevention trials and demonstrates the potential for premenopausal breast cancer prevention with progesterone receptor antagonists through stromal remodelling and luminal progenitor suppression.
19 April 2023 in Biomedical Physics & Engineering Express
Purpose. High breast density is associated with reduced efficacy of mammographic screening and increased risk of developing breast cancer. Accurate and reliable automated density estimates can be used for direct risk prediction and passing density related information to further predictive models. Expert reader assessments of density show a strong relationship to cancer risk but also inter-reader variation. The effect of label variability on model performance is important when considering how to utilise automated methods for both research and clinical purposes. Methods. We utilise subsets of images with density labels from the same 13 readers and 12 reader pairs, and train a deep transfer learning model which is used to assess how label variability affects the mapping from representation to prediction. We then create two end-to-end models: one that is trained on averaged labels across the reader pairs and the second that is trained using individual reader scores, with a novel alteration to the objective function. The combination of these two end-to-end models allows us to investigate the effect of label variability on the model representation formed. Results. We show that the trained mappings from representations to labels are altered considerably by the variability of reader scores. Training on labels with distribution variation removed causes the Spearman rank correlation coefficients to rise from 0.751 ± 0.002 to either 0.815 ± 0.026 when averaging across readers or 0.844 ± 0.002 when averaging across images. However, when we train different models to investigate the representation effect we see little difference, with Spearman rank correlation coefficients of 0.846 ± 0.006 and 0.850 ± 0.006 showing no statistically significant difference in the quality of the model representation with regard to density prediction. Conclusions. We show that the mapping between representation and mammographic density prediction is significantly affected by label variability. However, the effect of the label variability on the model representation is limited.
3 May 2024 in Biomedical Physics & Engineering Express
Purpose. To improve breast cancer risk prediction for young women, we have developed deep learning methods to estimate mammographic density from low dose mammograms taken at approximately 1/10th of the usual dose. We investigate the quality and reliability of the density scores produced on low dose mammograms focussing on how image resolution and levels of training affect the low dose predictions. Methods. Deep learning models are developed and tested, with two feature extraction methods and an end-to-end trained method, on five different resolutions of 15,290 standard dose and simulated low dose mammograms with known labels. The models are further tested on a dataset with 296 matching standard and real low dose images allowing performance on the low dose images to be ascertained. Results. Prediction quality on standard and simulated low dose images compared to labels is similar for all equivalent model training and image resolution versions. Increasing resolution results in improved performance of both feature extraction methods for standard and simulated low dose images, while the trained models show high performance across the resolutions. For the trained models the Spearman rank correlation coefficient between predictions of standard and low dose images at low resolution is 0.951 (0.937 to 0.960) and at the highest resolution 0.956 (0.942 to 0.965). If pairs of model predictions are averaged, similarity increases. Conclusions. Deep learning mammographic density predictions on low dose mammograms are highly correlated with standard dose equivalents for feature extraction and end-to-end approaches across multiple image resolutions. Deep learning models can reliably make high quality mammographic density predictions on low dose mammograms.
13 March 2025 in Biomedical Physics & Engineering Express
Background. Polygenic risk scores (PRS) summarise genetic information into a single number with clinical and research uses. Deep learning (DL) has revolutionised multiple fields, however, the impact of DL on PRSs has been less significant. We explore how DL can improve the generation of PRSs. Methods. We train DL models on known PRSs using UK Biobank data. We explore whether the models can recreate human programmed PRSs, including using a single model to generate multiple PRSs, and DL difficulties in PRS generation. We investigate how DL can compensate for missing data and constraints on performance. Results. We demonstrate almost perfect generation of multiple PRSs with little loss of performance with reduced quantity of training data. For an example set of missing SNPs the DL model produces predictions that enable separation of cases from population samples with an area under the receiver operating characteristic curve of 0.847 (95% CI: 0.828–0.864) compared to 0.798 (95% CI: 0.779–0.818) for the PRS. Conclusions. DL can accurately generate PRSs, including with one model for multiple PRSs. The models are transferable and have high longevity. With certain missing SNPs the DL models can improve on PRS generation; further improvements would likely require additional input data.
PurposeMammographic breast density is one of the strongest risk factors for cancer. Density assessed by radiologists using visual analogue scales has been shown to provide better risk predictions than other methods. Our purpose is to build automated models using deep learning and train on radiologist scores to make accurate and consistent predictions.ApproachWe used a dataset of almost 160,000 mammograms, each with two independent density scores made by expert medical practitioners. We used two pretrained deep networks and adapted them to produce feature vectors, which were then used for both linear and nonlinear regression to make density predictions. We also simulated an “optimal method,” which allowed us to compare the quality of our results with a simulated upper bound on performance.ResultsOur deep learning method produced estimates with a root mean squared error (RMSE) of 8.79 ± 0.21. The model estimates of cancer risk perform at a similar level to human experts, within uncertainty bounds. We made comparisons between different model variants and demonstrated the high level of consistency of the model predictions. Our modeled “optimal method” produced image predictions with a RMSE of between 7.98 and 8.90 for cranial caudal images.ConclusionWe demonstrated a deep learning framework based upon a transfer learning approach to make density estimates based on radiologists’ visual scores. Our approach requires modest computational resources and has the potential to be trained with limited quantities of data.