This study predicts ovarian cancer from cysts using TVUS and machine learning.
problem Early detection of ovarian cancer from cysts using TVUS screening.
method Employed Random Forest, KNN, and XGBoost machine learning techniques on PLCO dataset.
result Achieved high accuracy, recall, f1 score, and precision in predicting ovarian cancer.
Paper optimizes sparse feature selection for cancer detection using GSVP and SVM.
problem Sparse feature selection for cancer detection.
method Regularized GSVP with proximal gradient descent, feature selection via SVM.
result Near-perfect balanced accuracy with few selected features.
Develops a feature selection method for multi-view data with mixed types.
problem Challenges in feature selection for high-dimensional multi-view data with mixed data types.
method Block Randomized Adaptive Iterative Lasso (B-RAIL) combining randomized Lasso, adaptive weighting, and stability selection.
result Demonstrates effectiveness of B-RAIL in identifying biomarkers and novel candidates for ovarian cancer.
Paper integrates multiple data types for better cancer prediction.
problem How to effectively use diverse medical data for accurate cancer prediction.
method Multi-Kernel LS-SVM pipeline for integrative analysis of molecular and clinical data.
result Integration of various data types improves log-rank statistics and clinical status prediction.
New method recovers structured missing data in genomic studies.
problem Structured missingness in genomic data integration.
method Structured Matrix Completion (SMC) for efficient matrix recovery.
result Establishes optimal recovery rate and performs well in simulations and real data.
The objectives of this "perspective" paper are to review some recent advances in sparse feature selection for regression and classification, as well as compressed sensing, and to discuss how these might be used to develop tools to advance personalized cancer therapy. As an illustration of the possibilities, a new algor…
For mass spectra acquired from cancer patients by MALDI or SELDI techniques, automated discrimination between cancer types or stages has often been implemented by machine learnings. These techniques typically generate "black-box" classifiers, which are difficult to interpret biologically. We develop new and efficient s…
P-Net predicts patient outcomes using sample networks.
problem Predicting patient outcomes based on sample data.
method P-Net constructs a network of patients, ranking and classifying them based on phenotype.
result P-Net performs similarly to classical methods in predicting patient outcomes.
AI framework uses multi-omics data to personalize cancer treatment suggestions.
problem Leveraging AI for personalized cancer treatment based on complex patient characteristics.
method Modular machine learning framework trained on diverse multi-omics technologies.
result Superior performance in personalized counterfactual treatment suggestions.
New method for causal inference in survival outcomes using RDD.
problem Censoring in time-to-event analyses.
method Nonparametric approach with doubly robust censoring corrections.
result Higher efficiency and robustness to misspecification.
SWA selects important features from large data sets, controlling false discovery rate.
problem Feature selection in large regression data, especially scaling to big data and matching target FDR.
method Subsampling Winner algorithm using subsampling and scoring features.
result SWA controls actual FDR better than benchmark procedures and randomForest.
Due to advances in sensors, growing large and complex medical image data have the ability to visualize the pathological change in the cellular or even the molecular level or anatomical changes in tissues and organs. As a consequence, the medical images have the potential to enhance diagnosis of disease, prediction of c…
Estimates cost savings from early cancer diagnosis.
problem Improving early cancer diagnosis to reduce treatment costs.
method Combining published cancer treatment cost estimates by stage with incidence rates by stage at diagnosis, and extrapolating to other cancer sites.
result Estimates U.S. national annual treatment cost-savings from early cancer diagnosis in the trillions.
Model predicts anti-cancer drug responses using gene and molecular data.
problem Expensive and time-consuming cancer drug discovery and tailoring.
method Uses variational autoencoders and multi-layer perceptrons to encode gene expression and drug data.
result High average R2 of 0.83 and 0.845 in predicting drug responses for breast and pan-cancer cell lines, respectively. Study uses machine learning to predict heart failure in cancer patients.
problem Early detection of cancer patients at risk for cardiotoxicity.
method Examined four machine learning algorithms on 143,199 cancer patients.
result Gradient boosting model achieved best AUC score of 0.9077.
System accurately detects lung cancer from CT images.
problem Early and accurate detection of lung cancer.
method Developed algorithms using a dataset of CT images.
result Accuracy of 72.2% on test dataset.
Method extracts cancer signatures from genome data, reducing noise and variability.
problem Identifying stable cancer signatures from noisy genomic data.
method Applied statistical risk models from finance to cancer genome data, using NMF.
result Extracted signatures have lower variability and improved stability.
Deep learning models improve cancer detection and typing classification from gene expression data.
problem Challenges in establishing specificity for cancer diagnosis using gene expression data.
method Developed deep learning models using mRNA datasets for cancer detection and typing classification.
result Achieved 98% accuracy in cancer detection and 18 out of 32 cancer-typing classifications over 90% accuracy.
Study uses Apple ML to accurately detect and classify lung cancer.
problem Accurate diagnosis and sub-classification of non-small cell lung cancer.
method Evaluation of Apple Create ML module on histopathological images.
result 100% detection and successful subclassification of non-small cell lung cancer.
Study predicts 10-year survival rates for breast cancer patients.
problem Predicting long-term survival of breast cancer patients.
method Machine learning approaches to assess survival rates.
result Improved accuracy in predicting 10-year survival.
New *K-means method finds cancer signatures without NMF.
problem Identifying cancer signatures from genome data.
method Applying *K-means clustering to 1,389 cancer samples.
result 3 cancers lack cluster-like structures, 2 have high correlations.
Cancer patients admitted to ICU had improved survival over 10 years.
problem To assess changes in survival of cancer patients admitted to ICU over 10 years.
method Retrospective analysis of MIMIC-III database, adjusted for confounders using logistic regression.
result Cancer patients had significantly lower 28-day and 1-year mortality rates over 10 years.
Paper identifies key CpG methylation sites for breast cancer.
problem Early detection and treatment of breast cancer.
method Used machine learning on TCGA dataset to classify cancer vs. non-cancer samples.
result Reduced model with 25 key CpG sites achieves over 94% accuracy.
Noise-filtering improves cancer drug sensitivity prediction.
problem Improving accuracy of cancer drug sensitivity predictions.
method Integrates numerical linear algebra and information retrieval techniques to filter out noisy cancer cell lines.
result Our approach yields the highest AUC on clinical trial data.
Data mining techniques predict breast cancer types with high accuracy.
problem Early detection of breast cancer to reduce mortality rates.
method Twelve classification algorithms applied to the Breast Cancer Wisconsin dataset.
result High accuracy in predicting malignant and benign breast cancer.
Study examines perceptions and attitudes about breast cancer on Twitter.
problem Understanding public perceptions and attitudes towards breast cancer on social media.
method Identified and collected tweets, used topic modeling and sentiment analysis.
result Identified themes and quantified users' perceptions and emotions about breast cancer.
Machine learning clusters mutations in cancer exomes, improving diagnostic speed and cost.
problem Extracting stable mutation structures from cancer exome data for early diagnostics.
method Statistically deterministic machine learning algorithm *K-means applied to exome samples.
result Majority of cancer types exhibit stable mutation clustering, while NMF methods are unstable.
Deep learning predicts breast cancer with high accuracy from patient data.
problem Early detection of breast cancer from patient data.
method Feature selection and k-fold Monte Carlo cross-validation using deep learning.
result Deep learning model effectively distinguishes between cancer and healthy patients.
Machine learning accurately diagnoses cancer from whole genome sequencing data.
problem Accurate cancer diagnosis at all stages.
method Novel MLAC (Machine Learning Against Cancer) method using next-gen RNA sequencing.
result Perfect precision, sensitivity, and specificity achieved for most tumor types.
Neural networks improve cancer risk prediction from family history data.
problem Improving cancer risk prediction from family history data using machine learning.
method Developed and trained neural network models on large pedigrees to predict hereditary cancers.
result Neural networks can achieve nearly optimal prediction performance and outperform traditional models in misreported data.
Machine learning detects metastatic breast cancer cases from linked EMR and cancer registry data.
problem Lack of metastatic recurrence data in cancer registries and EMRs.
method Semi-supervised machine learning on linked EMR and CCR data.
result Model achieved high accuracy in detecting metastatic breast cancer cases.
L-Perceptron improves breast cancer diagnosis and survival prediction.
problem Improving early prognosis and survival prediction rates for breast cancer.
method Proposes a novel type of perceptron (L-Perceptron) for better accuracy and sensitivity.
result Achieves 97.42% and 98.73% accuracy and sensitivity in Wisconsin Breast Cancer dataset.
System automates identification of cancer drug repurposing from PubMed.
problem Manual extraction of cancer drug repurposing evidence from scientific publications is infeasible.
method NLP pipeline including querying, filtering, entity extraction, classification, and study type classification.
result Automated system extracts cancer drug repurposing evidence from PubMed abstracts.
Deep neural networks improve cancer nuclei classification and survival analysis.
problem Accurate cell nuclei classification for cancer diagnosis and survival analysis.
method Implementation and evaluation of deep neural network models and ensembles for nuclei classification in RCC and PCa.
result Convolutional neural network system based on residual learning significantly improves cell nuclei classification.
PathologyGAN learns deep representations of cancer tissue images.
problem Limited high-quality labels for cancer tissue images.
method Developed a GAN framework for unsupervised learning of cancer tissue phenotypes.
result Generated high-quality images with interpretable latent space.
Study identifies biomarkers for lung cancer in female non-smokers.
problem Identifying prognostic biomarkers for stage III NSCLC in non-smoking females.
method Gene expression profiling and XGBoost machine learning algorithm.
result Top biomarkers validated in literature, with AUC score of 0.835.
Deep autoencoder predicts cancer types from DNA methylation patterns.
problem Differentiating cancer types based on DNA methylation states.
method Deep learning system with CpG island state classification and statistical methods.
result Overall Sensitivity of 88.24%, Specificity of 83.33%, Accuracy of 84.75%.
Deep CNN model improves breast cancer screening exam classification.
problem Improving accuracy in breast cancer screening exam classification.
method Localization-based deep CNN trained on 200,000 exams.
result AUC of 0.919 in predicting malignancy, reducing error rate by 23%.
Modeling correlated mutations in cancer for personalized treatment.
problem Identifying mutations for personalized cancer therapy in heterogeneous profiles.
method Proposed correlated zero-inflated negative binomial process with mixed beta-Bernoulli and variational inference.
result Identified biologically relevant correlations between somatic mutations.
Network Elastic Net identifies smoking-specific gene expression for lung cancer prognosis.
problem Identifying smoking-specific gene expression biomarkers in lung cancer prognosis.
method Introduces Network Elastic Net, a method that clusters and regresses on graphs based on smoking behavior.
result Shows efficacy of clusters in identifying cancer stages using gene expression and smoking behavior.
The study improves colorectal cancer survivability prediction by considering ethnicity.
problem Improving colorectal cancer survivability prediction using machine learning.
method Machine learning techniques applied to SEER cancer incidence database, comparing different ethnicities.
result Models perform better on single-ethnicity populations and provide different feature importance rankings.
Automates deep learning model development for cancer data.
problem Manual design of high-performing deep learning models for cancer data is time-consuming and requires expertise.
method Reinforcement-learning-based neural architecture search with custom building blocks.
result Automated discovery of deep neural network architectures with similar or higher accuracy.
Tree model predicts breast cancer in older women.
problem Breast cancer prediction in older women with DCIS grades.
method Tree augmented naive Bayesian network trained on clinical data.
result Biopsy threshold recommendation of >2% for older women.
BIDIFAC+ factorizes linked matrices for cancer studies.
problem Integrating multiple omics platforms across various cancer types.
method Flexible approach to simultaneous factorization and decomposition of linked matrices using BIDIFAC+.
result Identifies shared and specific modes of variability across multiple omics platforms and cancer types.
Deep learning improves prostate cancer diagnosis using MRI.
problem Improving accuracy in prostate cancer diagnosis.
method Developed XmasNet, a deep learning architecture based on convolutional neural networks, trained on 3D multiparametric MRI data.
result XmasNet outperformed traditional machine learning models and achieved the second highest AUC (0.84) in the PROSTATEx challenge.
Paper proposes scalable method for analyzing multi-omic data.
problem Integrating high-dimensional multi-omic data for cancer subtyping.
method Mixed graphical model approach using Birth-Death MCMC algorithm.
result Our method outperforms LASSO and standard BDMCMC in computational efficiency and model selection accuracy.
Model learns cancer tissue images onto a low-dimensional space revealing tissue characteristics.
problem Improving cancer diagnosis through high-fidelity digital pathology.
method Deep generative model using PathologyGAN to map real images onto a latent space.
result Latent space encodes morphological characteristics and reveals distinct tissue clusters.
Machine learning can predict cancer with 100% accuracy on a dataset.
problem Accuracy of cancer predictions using machine learning.
method Extensive experiments on the Wisconsin Diagnostic Breast Cancer dataset.
result Machine learning algorithms can be easily misled to achieve 100% accuracy.