Artificial Intelligence in Early Breast Cancer Screening

Accurate and computationally efficient classification systems are essential for the early detection of breast cancer, particularly when dealing with complex and high-dimensional medical datasets. Traditional machine learning models often face limitations in capturing intricate nonlinear relationships inherent in such data, potentially compromising diagnostic performance. In this study, we introduce QBG-QCNN, a Quantum-enhanced framework named Boruta-GA optimized Quantum Convolutional Neural Network, designed for breast cancer classification. The model is trained on the Breast Cancer Wisconsin (Diagnostic) Dataset, which contains 30 numerical features extracted from fine needle aspiration (FNA) images of breast tissue samples. To reduce dimensionality while preserving critical diagnostic information, a hybrid Boruta-GA feature selection strategy is applied to extract key features such as radius_mean, texture_mean, area_mean, and concavity_mean. These selected features are then encoded into a 4-qubit quantum circuit using advanced quantum feature maps ZZFeatureMap, RealAmplitudes, and EfficientSU2, eliminating the need for manual feature engineering. The encoded quantum data is processed through a QCNN that incorporates quantum convolution, pooling, and parameterized ansatz layers, leveraging quantum entanglement and parallelism for more efficient learning. Implemented using PennyLane and IBM Qiskit, and optimized with the COBYLA, the model achieves outstanding performance: 94.3% accuracy, 95.2% precision, 94.6% recall, and a 93.0% F1-score. These results significantly outperform those of classical CNNs, standard QNNs, and other hybrid models. In conclusion, QBG-QCNN demonstrates that quantum machine learning, when integrated with intelligent feature selection, offers a powerful, scalable, and interpretable solution for early-stage breast cancer diagnosis. Future research will extend this framework to multi-modal datasets and real-device deployment on real quantum devices under noise constraints.

<p>Breast cancer is ranked as a significant cause of mortality among females globally. Its complex nature poses principal challenges for physicians and researchers for rapid diagnosis and prognosis. Hence, machine learning algorithms are employed to forecast and identify diseases. This study discusses the comparative analysis of seven machine learning models, e.g., logistic regression (LR), support vector machine (SVM), k-nearest neighbor classifier (KNN), decision tree classifier (DT), random forest classifier (RF), Naïve Bayes (NB), and artificial neural network (ANN) to predict breast cancer using Wisconsin breast cancer and breast cancer datasets. In the Wisconsin breast cancer dataset, KNN depicted 99% accuracy, followed by RF (98%), SVM (96%), NB (96%), LR (96%), ANN (93%), and DT (92%). On the contrary, in the breast cancer (BC) dataset, the highest accuracy was achieved by LR at 83%, and the lowest was achieved by DT (65%), which depicted that the numeric dataset WBC has better accuracy than the breast cancer dataset.</p>

Background Earlier detection of breast cancer through mammography screening has reduced disease-specific mortality; however, confounding issues such as technical challenges, breast density, and tumor size can result in false negatives and ultimately later stage diagnosis. Next generation liquid biopsy has the potential to complement mammography and enable earlier detection for more women. We have previously demonstrated high sensitivity and specificity for early detection of invasive breast cancer (IBC) by utilizing a novel category of cancer-associated small RNAs, termed orphan noncoding RNAs (oncRNAs), through a liquid biopsy platform. Here, we further improve the ability to detect breast cancer in a larger, multi-source cohort through an AI-driven approach and demonstrate potential for detection of ductal carcinoma in-situ (DCIS). Methods We utilized The Cancer Genome Atlas (TCGA) small RNA-seq database to discover a library of 20,538 oncRNAs, through a female-specific analysis, that were significantly enriched among 1,103 breast tumors compared to 349 normal tissue samples spanning multiple tissue sites. The diagnostic performance of these oncRNAs were assessed in an independent cohort of archived serum samples from 663 female individuals, sourced from Indivumed (Hamburg, Germany), Proteogenex (Inglewood, CA), and MT Group (Los Angeles, CA), including 279 breast cancer patients of various stages (221 IBC and 58 DCIS; mean age: 57.0 ± 13.8 years; ever-smoker: 25.8%) and 304 age-matched controls (mean age: 58.5 ± 13.9 years; ever-smoker: 23.4%) without breast cancer. All samples were collected between 2010–2022 at time of diagnosis for breast cancer patients. We sequenced the small RNA content of these samples at an average depth of 25.28 ± 9.37 million 50-bp single-end reads. We detected 18,025 (87.8%) unique breast cancer-specific oncRNA species within at least one sample from the study cohort. We then trained a generative AI model using 5-fold cross-validation to predict cancer status for all samples. Results Our oncRNA-based model achieved an overall AUC of 0.95 (95% CI, 0.93–0.97) for prediction of IBC versus cancer-free controls with a sensitivity of 0.87 (0.82–0.91) at 90% specificity. We observed high sensitivities, also at 90% specificity, across all tumor stages and tumor sizes (Table 1). Sensitivities for the earliest stage and smallest tumor size were 0.87 (0.78–0.93) and 0.81 (0.61–0.93) for Stage I (n=83) and T1a–b ( >1mm to ≤10mm; n=26), respectively. Additionally, in a small single-source cohort, we also saw high model accuracy and sensitivity for DCIS, which we aim to confirm in additional cohorts. While our overall cancer cohort primarily consisted of individuals with luminal breast cancer, our model had high sensitivities across all breast cancer subtypes at 0.90 (0.84–0.94), 0.73 (0.59–0.85), and 0.86 (0.42–1.0) for luminal (n=181), HER2 positive (n=49), and triple negative (n=7), respectively. Conclusions We further demonstrate the potential utility of oncRNAs as a blood-based biomarker using an AI algorithm for sensitive and accurate early detection of breast cancer in a large cohort. Additionally, we have shown that this oncRNA-based assay performs well in detecting small, early-stage invasive breast tumors, with potential to detect precursors of breast cancer. Table 1: Model sensitivity in breast cancer by tumor stage and size For each tumor stage and size, as defined by the AJCC 7th Edition breast cancer staging system, sensitivity and 95% Pearson-Clopper confidence intervals (CI) are reported at 90% specificity for the number of samples (N). Citation Format: Noura Tbeileh, Taylor Cavazos, Mehran Karimzadeh, Jeffrey Wang, Alice Huang, Dung Ngoc Lam, Seda Kilinc, Jieyang Wang, Xuan Zhao, Andy Pohl, Helen Li, Lisa Fish, Kimberly Chau, Marra Francis, Lee Schwartzberg, Patrick Arensdorf, Hani Goodarzi, Fereydoun Hormozdiari, Babak Alipanahi. Cell-free orphan noncoding RNAs and AI enable early detection of invasive breast cancer and ductal carcinoma in-situ [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-13-08.

The breast cancer diagnosis is one of the challenging tasks of medical field. Especially, the breast cancer diagnosis among younger women (under 40 years old) is more complicated, because their breast tissue is generally denser than the older women. The Breast Cancer Wisconsin image dataset contains two classes: (i) Benign (Minority class), (ii) Malignant (Majority class). The imbalanced class distribution leads to a deterioration in the classifier model performance owing to the biased classification towards the majority class. Therefore, in this article, an improved generative adversarial network (I-GAN) is proposed to overcome the class imbalance problem. Here, the proposed method is the consolidation of deep convolutional generative adversarial network (DCIGAN) and modified convolutional neural network, (MCNN), therefore it is known as DCIGAN-MCNN method. First, the DCIGAN is utilized for balancing the dataset by generating more samples in the training dataset. Then, this training dataset based the classification of Breast cancer is developed using the modified convolutional neural network. The proposed method is executed in MATLAB. The performance analysis are carried out in Breast Cancer Wisconsin (Prognostic) Data Set provides Maximum Geometry Mean (MGM) as 24.058%, 9.582%, Matthews’s correlation coefficient (MCC) as 78.623%, 30.357% higher than the existing methods, like CI-BC-RK-SVM, CI-BC-GA, CI-BC-DC-CNN, CI-BC-RF and CI-BC-BMIC-Net respectively. Finally, the simulation results prove that the proposed method can be able to find the optimal solutions efficiently and accurately.

Breast cancer is the most frequently occurring cancer disease in women. It is reported almost 14% of cancers in Indian women are breast cancer. It becomes very crucial to predict breast cancer earlier to minimize the deaths. This research article helps to predict breast cancer earlier and reduce the immature deaths of women in India. In this paper, the authors have used the Logistic Regression method to classify the disease. The authors simulate the results using logistic regression with 10-fold cross-validations and with a different train-test split of the dataset. The 10-fold cross validations display its potential with almost 94% performance in the research paper. With all features and 90-10 , 80-20,50-50, 66-34 splits, and 10-fold cross-validations the authors achieve 96% accuracy. we have used different accuracy measures like accuracy, sensitivity, specificity, and kappa statistics to get the novelty of the model. In this study, the authors use the Wisconsin (Diagnostic) Data Set for Breast Cancer, Created by Dr. William H. Wolberg, General Surgery Dept., University of Wisconsin, Clinical Science Centre, Madison, WI 53792 wolberg@eagle.surgery.wisc.edu available at the UCI ML Repository website.

In this paper four machine learning algorithms are compared in order to predict if a cell nucleus is benign or malignant using the Breast Cancer Wisconsin (Diagnostic) Data Set. The algorithms are K-Nearest Neighbours, Classification and Regression Trees (CART), Naive Bayes and Support Vector Machines with Radial Basis Function Kernel. Data visualization and Pre-Processing using PCA will help in the understanding and the preparation of the dataset for the training phase while parameter tuning will determine the optimal parameter for every model using R as programming language. Also, 10-fold Cross Validation is used as a resampling method after comparing it with Bootstrapping, as it is the most efficient out of the two. In the end, our comparison shows that the machine learning model that marked the highest Accuracy is the one that is trained using K Nearest Neighbours. Nowadays, one of the most common forms of cancer among women is breast cancer with more than one million cases and nearly 600,000 deaths occurring worldwide annually [1]. It is the second leading cause of death among women and thus it must be detected at an early stage in order not to become fatal [2]. Thus, the importance of diagnosing if a biopsied cell is benign or malignant is vital. However, this process is quite complicated as it involves several stages of gathering and analysing samples with many variables, making the final diagnosis a demanding and timely procedure. The rapid growth of Artificial Intelligence and Machine learning and their implementation in Medicine give us a new perspective in the way we process and analyse medical data. Medical experts can use Data Mining techniques and improve their decision making by extracting useful information from massive amounts of data.

Breast cancer is one of the most common forms of cancer that is diagnosed in most women and some rare cases even men. In recent years, breast cancer survival rates have increased significantly, due to factors such as earlier detection. Treatment for breast cancer largely depends on identifying the type of mass of tissue formed, which is known as a tumor. If normal cells grow in an uncontrollable manner the tumor is called benign (non-cancerous). But, if the cells’ growth is out of control and their behavior is abnormal, then the tumor is called malignant (cancerous). During the invasive, (i.e. curable) stage of cancer, only 10–15% part of the breast contains cancerous cells. Therefore, it is difficult to diagnose it using mammography. However, the development of machine learning techniques has allowed early detection of breast cancer in clinical trials. This paper presents a comparative study of different machine learning techniques for detecting breast cancer. It also presents an auto-encoder model that performs breast cancer detection in an unsupervised manner. It attempts to identify a compact representation of features that are strongly related to breast cancer. The techniques are tested on the Breast Cancer Wisconsin (Diagnostic) Dataset, which is publicly available on Kaggle. The auto-encoder beats its competitors with a precision and recall of 98.4%.KeywordsBreast cancerSupervised learningUnsupervised learningAuto-encoder

While there are many factors which could contribute to the occurrence of breast cancer, it is very difficult to attribute the exact environmental and other factors contributing to it, but still it has significance in determining the occurrence of cancer. Using machine learning techniques and regular diagnosis information, we can achieve our goal of assessing the risk of occurrence of breast cancer. Cancer data sets contain many attributes of patient information, but not every feature is relevant in predicting cancer. Feature selection techniques are useful in such scenarios for retaining the relevant feature set. In this paper we are doing a comparative study of the effect of feature selection techniques on the accuracies given by existing machine learning algorithms. For this purpose we have considered the following machine learning algorithms - Logistic Regression, Naive Bayes and Random Forest. The following feature selection techniques have been considered - Sequential Forward Feature Selection, Recursive Feature Elimination, f-test and correlation.The publicly available Breast Cancer Wisconsin (Diagnostic) Data Sets from UCI Repository have been used in this work. The results show that random forest algorithm gives the highest accuracy with feature selection. Furthermore f-test gives better results for the smaller dataset and Sequential Forward Selection for the larger dataset.

Data reduction is an indispensable part of pattern classification processes in many cases. If the number of samples is excessive, sample reduction or data reduction algorithms can be used for an effective processing time and reliable successive results. Many methods have been used for data reduction. Fuzzy c-means is one of these methods and it is widely used in such applications as clustering algorithms. In this study, we applied a different clustering algorithm, an artificial immune system (AIS), for the data reduction process. We realized the performance evaluation experiments on the standard Chainlink and Iris datasets, while the main application was conducted using the Wisconsin Breast Cancer and Pima Indian datasets, which were taken from the University of California, Irvine Machine Learning Repository. For these datasets, the performance of the AIS in the data reduction process was compared with the fuzzy c-means clustering algorithm, in which a multilayer perceptron artificial neural network was used as a classifier after the data reduction processes. The obtained results show that the maximum classification accuracies were obtained as 73.96 % for the Pima Indian Diabetes dataset and 97.80% for the Wisconsin Breast Cancer dataset with the AIS and the compression rates were 80% and 40% for these results. For fuzzy c-means clustering, however, the aforementioned accuracies were obtained as 63% and 86.69% for the Pima Indian Diabetes and Wisconsin Breast Cancer datasets, respectively. Moreover, the compression rates for these results for fuzzy c-means were 90% and 70%. When the mean classification accuracy values over the experimented compression rates were taken into consideration, the AIS reached a mean classification accuracy of 70.07% for the Pima Indian Diabetes dataset, while 47.64% was obtained by fuzzy c-means for this dataset. For the Wisconsin Breast Cancer dataset, however, the mean classification accuracies of the AIS and fuzzy c-means methods were recorded as 94.90% and 75.43%, respectively.

To explore the early detection of breast cancer by ultrasonic imaging and thermal tomography of luciferase or green fluorescent protein (GFP)-labeled MDA-MB-231 breast cancer cell line-xenografts in nude mice. Fluorescence-tagged lentiviral vectors were transfected into the triple-negative breast cancer cell line MDA-MB-231. These cells were implanted either subcutaneously under the right breast pad or intravenously into the tail vein of nude BALB/C mice. Thermal tomography and ultrasound imaging were used to detect tumor formation and to monitor tumor growth and metastasis in vivo. Triple negative breast cancer cell line-xenografts were used to successfully construct an orthotopic nude mice model of breast cancer metastasis in the peritoneum. Thermal tomography and ultrasound imaging were used together to detect small tumors. Thermal tomography imaging detected small tumors earlier than ultrasound imaging. Thermal tomography can be used to monitor changes in tumor growth and detect abnormal tissue. Therefore, it can serve as a convenient,rapid,sensitive, and reliable technique for early screening of human breast cancer.

Early detection of breast cancer significantly improves survival rates; however, existing screening modalities such as mammography are limited by radiation exposure, discomfort, reduced sensitivity in dense breasts, and poor accessibility in low‑resource settings. This paper proposes a novel, non‑invasive, radiation‑free IoT‑enabled wearable patch integrated with thermal sensors and complementary biosensors to enable early breast cancer screening through continuous physiological monitoring. The patch wirelessly connects to a smartphone and cloud platform, where artificial intelligence (AI) models analyze longitudinal data to identify abnormal thermal asymmetry, vascular heat dispersion, and biosignal patterns indicative of tumor‑associated angiogenesis and metabolic activity. The proposed system acts as a first‑line screening and triage tool, enabling frequent, home‑based monitoring and timely referral for diagnostic imaging. System architecture, sensing methodology, AI pipeline, and potential clinical impact are presented, demonstrating the feasibility of scalable, personalized breast cancer screening using wearable technology.

Introduction: Women with dense breast tissue face a two-fold higher risk of breast cancer, yet mammography sensitivity is as low as 30% in this group, underscoring the need for more sensitive and accessible screening methods. Liquid-based biopsies for early breast cancer detection are emerging but currently remain out of reach for clinical use. Recent data on nucleotide assessment from plasma in breast cancer have been mixed with 87% sensitivity for late-stage disease (stage 3-4) but only 20% sensitivity for early-stage disease (stage 1-2). Proteomics is an exciting area for early cancer screening where improvements in sample preparation and equipment have enabled major advances in early cancer detection. This is particularly true for a deep proteome assay as the detection of proteins 8-9 orders of magnitude lower in abundance than common plasma proteins can be identified. Objective: This study evaluates the presence of breast cancer using label-free shotgun mass spectrometry-based proteomics on less than 1 ml of plasma in >1,100 women classified as either healthy or as newly diagnosed breast cancer patients with a focus on early-stage disease (stages 0-2). Methods: 363 banked blood samples from multiple sources were used to establish a training set consisting of 156 stage 0-4 treatment-naive breast cancer patients, 46 benign breast conditions, and 161 healthy controls. Samples were age and demographic matched between healthy, benign, and breast cancer cohorts. Samples were processed in a blinded manner using an automated and novel deep proteomic platform developed by Astrin Biosciences. This platform combines label-free quantitative mass spectrometry and a proprietary breast cancer-specific spectral library to extract features used to train a classifier to identify breast cancer patients. Blinded validation of this AI classifier was evaluated on the remaining 788 patient samples (healthy, n=374; benign, n=81; breast cancer, n=333) to confirm performance characteristics. Results: Preliminary analyses from our training set of deep proteomic profiling identified over 8,600 proteins in total and up to 7,500 proteins per patient (median 6,926 proteins) with a dynamic range spanning over 8 orders of magnitude. Differentially expressed proteins in early-stage breast cancer patients from healthy patient controls were also observed. Our AI classifier, using a leave-one-out cross-validation (LOOCV) approach on the interim cohort, separated healthy controls from breast cancer patients with an ROC of >95% and a specificity and sensitivity greater than 85% across all stages. Conclusion: We have developed a highly sensitive blood-based assay that utilizes deep proteomic profiling to identify distinctive cancer specific signatures in women who are undergoing screening for breast cancer. This work enables us to develop a protein-based classifier from plasma for early detection of breast cancer. Impact on screening strategies: This assay advances clinical diagnostics by identifying proteomic markers for early detection of breast cancer. This innovative approach enhances screening strategies for women with dense breasts who are at average or high-risk based on family history, specific genetic mutations such as BRCA1/2, race, or other factors. Acknowledgements: We thank all of the members at Astrin Biosciences who have contributed endless hours dedicated to ending cancer. We are also grateful to all the donors who generously contributed blood samples to this study. Citation Format: A. Horrmann, Y. Travadi, G. Schaap, K. Mallery, K. J. Kamalanathan, N. Bristow, C. Galeano-Garces, S. Y. Bae, A. Hesch, C. Rungkittikhun, B. R. Konety, J. M. Drake. Deep Proteomics and AI Classifier for Early Breast Cancer Detection [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PD5-03.

Introduction: The use of social media by healthcare professionals to create disease awareness has been increased worldwide. It is easily available and accessible to the larger number of populations and hence helps in bridging the gap between doctor and patient. Breast cancer is the leading cause of cancer among women in India and the main reason for higher mortality among them is the late presentation to the surgeons, owing to various myths and social and physical taboos. We conducted a study to assess the awareness of breast cancer among the women in general population and also to check the effectiveness of WhatsApp in improving women's knowledge regarding the risk reduction, myths, early detection and treatment of breast cancer. The study was a pre-post health educational study including 146 wom Methods: en. Multiple WhatsApp groups were created to share information regarding breast cancer awareness, early detection, risk factors, common myths, self-breast examination and treatment of breast cancer in the form of audio, video, text and images. The group was active for 4-5 days as an interactive session. Data were collected through google forms. We noticed a signicant improvement in the knowledge of women regard Results: ing the common myths regarding breast cancer and the symptoms of breast cancer. There was signicant increase in performance of self-breast examination from 14.4% to 84.6% (p- <0.001). Most of the women found the video of self-breast examination useful. The commonly Conclusion: used social networking site like WhatsApp can be used as a tool to assess the awareness of breast cancer among women. This also helped in exchange of the knowledge among women, which in turn helped in the risk reduction and early detection of breast cancer. This app is a useful tool that can be incorporated into health education strategies that focus on breast cancer.

Radiologic techniques remain the main method for early detection for breast cancer and are critical to achieve a favorable outcome from cancer. However, more sensitive detection methods to complement radiologic techniques are needed to enhance early detection and treatment strategies. Using our recently established culturing method that allows propagation of normal and cancerous breast epithelial cells of luminal origin, flow cytometry characterization, and genomic sequencing, we show that cancer cells can be detected in breast milk. Cells derived from milk from the breast with cancer were enriched for CD49f+/EpCAM-, CD44+/CD24-, and CD271+ cancer stem-like cells (CSC). These CSCs carried mutations within the cytoplasmic retention domain of HDAC6, stop/gain insertion in MORF4L1, and deletion mutations within SWI/SNF complex component SMARCC2. CSCs were sensitive to HDAC6 inhibitors, BET bromodomain inhibitors, and EZH2 inhibitors, as mutations in SWI/SNF complex components are known to increase sensitivity to these drugs. Among cells derived from breast milk of additional ten women not known to have breast cancer, two of them contained cells that were enriched for the CSC phenotype and carried mutations in NF1 or KMT2D, which are frequently mutated in breast cancer. Breast milk-derived cells with NF1 mutations also carried copy-number variations in CDKN2C, PTEN, and REL genes. The approach described here may enable rapid cancer cell characterization including driver mutation detection and therapeutic screening for pregnancy/postpartum breast cancers. Furthermore, this method can be developed as a surveillance or early detection tool for women at high risk for developing breast cancer. SIGNIFICANCE: These findings describe how a simple method for characterization of cancer cells in pregnancy and postpartum breast cancer can be exploited as a surveillance tool for women at risk of developing breast cancer.

Currently, as governments across the world attempt to halt the spread of COVID-19, many of their efforts are being hampered by shortages of crucial testing kits. Even in many developed countries, the health system has come to the point of standstill due to the increasing demand for intensive care units. One study reveals that 40,000 women die in a year from breast cancer, which is one woman every 13 minutes dying from this disease every day. Breast cancer remains the number one form of cancer that women are diagnosed with around the world. Any hospital in the world will have a multidisciplinary team of experts including radiologists, oncologists, surgeons, and other respiratory physicians and consultants. This multidisciplinary team uses a wide variety of data taken from CT scans, x-rays, PET scans, biopsies, clinical examination, imaging mammography, ultrasound (sonography), pathology biopsy, and fine-needle aspiration to identify whether a patient has cancer. Individuals with cancer, particularly those who are receiving systemic anticancer treatments, have been postulated to be at increased risk of mortality from COVID-19. Therefore, it is the need of an hour to implement an early detection of cancer that will prevent patients from COVID-19 and due treatment will be provided. Breast cancer that is detected early is far easier to cure, but at the same time it is difficult to detect. An accurate breast cancer staging is an essential task performed by pathologists (experts who examine the laboratory samples) to inform clinical management for the appropriate treatment. For instance, evaluation of the extent of cancer spread by histopathologic analysis or microscopic examination is an important part of breast cancer staging, and the process involves the decision-making of human behavior with time-consuming and costly sample examination and a period of more than one month to have breast cancer type, stage, etc., to decide the appropriate therapy to patients by the time it can affect the results of therapeutic treatments and survival of patients. The application of artificial intelligence (AI) machine learning technology with deep learning algorithms to whole-slide pathology images can potentially improve the diagnostic accuracy of breast cancer at a very early stage. I have developed a simple machine-learning model using the Python programming language and neural network, inspired by the structure of the human brain, with 12 attributes, out of which 10 real-valued features from each cell nucleus obtained from the Wisconsin diagnostic breast cancer dataset available in public domain for the education capacity that explains about the stage of breast cancer M (Malignant) and B (Benign). This model predicts the presence of breast cancer with an accuracy of 98.8% at a very early stage. The continuous development of technology in the medical field will save countless lives and the overall quality of human life will continue to improve over time. Citation Format: Subash Kumar. Early detection of breast cancer during the Covid-19 pandemic using artificial intelligence neural networks [abstract]. In: Proceedings of the AACR Virtual Meeting: COVID-19 and Cancer; 2020 Jul 20-22. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(18_Suppl):Abstract nr PO-003.