Technological advances have ushered in a new multi-omics era that has elucidated more granular information on cell types and functions at single-cell resolution. Integrating morphology as a multi-dimensional readout of cell identity, state, and function is an area of broad interest, since it has been historically difficult to quantify morphological properties. Here, we apply artificial intelligence (AI) and multi-dimensional morphology to characterize a variety of human cell types and states using the Deepcell platform. Captured brightfield images of cells flowing inside a microfluidic channel are analyzed by AI models in real-time and cells of interest can be sorted for downstream molecular or functional analyses. We developed an AI model, termed ‘Human Foundation Model’ (HFM), for broad use that combines machine learning with computer vision for multi-dimensional morphology analysis to provide interpretable features on a broad range of human cells and diverse sample types. The HFM training and validation datasets, comprising 1.18 million and 4.05 million cell images, respectively, were chosen to capture a comprehensive range of visual features and cover a wide spectrum of biological diversity. Samples included 6μm polystyrene beads (as controls) and multiple immune and carcinoma cell lines. Subsets of these images were labeled for specific cell features then included in the supervised portion of the model training. We apply the HFM as part of the Deepcell system to identify and characterize the morphological heterogeneity of tumor cells in different types of cancers such as melanoma, non-small cell lung carcinoma, and malignant pleural effusions. We further demonstrate the ability of the platform to distinguish different cell types (epithelial, stromal, immune, and endothelial cells) commonly found in tumors that typically require complex antibody panels. With the Deepcell platform and the HFM, we determined cell identities using morphological features associated with distinct processes characterized by granules, vesicles, cell size, pigmentation and others. Furthermore, sorted cells of interest are unlabeled and can be retrieved for further analysis. This AI-powered technology can be applied to diverse areas including drug screenings/mechanism of action, tumor biology, and functional genomics. Citation Format: Stephane C. Boutet, Anastasia Mavropoulos, Andreja Jovic, Jeanette Mei, Nianzhen Li, Kiran Saini, Senzeyu Zhang, Chassidy Johnson, Vivian Lu, Ryan Carelli, Kevin B. Jacobs, Mahyar Salek, Maddison Masaeli. A broad-use deep learning model based on multi-dimensional morphology to identify and characterize tumor cell heterogeneity. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5381.

Understanding the tumor microenvironment and detecting rare circulating tumor cells from blood are two major challenges faced by cancer biologists and oncologists. Both require high sensitivity and accuracy in classifying and isolating single cells in a complex and heterogeneous environment. Classical cell classification and sorting techniques are limited by their reliance on pre-selected cell biomarkers or physical characteristics. Recent breakthroughs in machine learning have achieved unprecedented accuracy across a wide range of image classification problems. We have developed a platform that combines high-resolution imaging of unlabeled cells in microfluidic flow with real-time deep neural network (DNN) based classification and sorting. The DNN classifier was trained on more than 25 million high-resolution cell images of multiple types imaged on the platform. Our model was trained to discriminate among multiple cell classes, including immune cell subtypes, non-small-cell lung cancer cells (NSCLC), hepatocellular carcinomas (HCC), and stromal cells (including endothelial, epithelial, fibroblasts, smooth muscle cells). We then assessed model performance on a separate validation set of cell images, including cell lines not used in the training data. Our classifier accurately identifies NSCLC and HCC against a background of blood cells with an area under the ROC curve (AUC) of > 0.999. In addition we demonstrate the enrichment of NSCLC cells from spike-in mixtures with WBCs or whole blood at concentrations as low as 1:100,000, achieving an enrichment of > 25,000x on multiple cell lines. Using dissociated lung cancer tissue, we demonstrate that our label-free classification of tumor cells closely matches results from both standard flow cytometer analysis and single cell RNA sequencing. Additionally we were able to enrich the tumor cell fraction from dissociated tumor tissue thereby improving the sensitivity of mutation detection and enabling refined downstream single-cell genomic analysis. This work demonstrates that deep learning using high-resolution cell images collected at scale can achieve a high classification accuracy and can enable the label-free isolation of rare cells of interest for a wide range of applications. This system can be used to analyze tumor biopsies and liquid biopsies and has the potential to enable the study of tumor cells and tumor microenvironment with novel dimension and insight. Citation Format: Mahyar Salek, Hou-pu Chou, Prashast Khandelwal, Krishna P. Pant, Thomas J. Musci, Nianzhen Li, Christina Chang, Andreja Jovic, Esther Lee, Stephanie Huang, Jeff Walker, Phuc Nguyen, Kiran Saini, Jeanette Mei, Quillan F. Smith, Maddison Masaeli. Deep learning enables label-free profiling of the tumor microenvironment and enrichment of rare cancer cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 188.

No abstract