Circulating Tumor Cell Identification Based on Deep Learning

Quantitative image analysis pipeline for detecting circulating hybrid cells in immunofluorescence images with human-level accuracy

Circulating hybrid cells (CHCs) are a newly discovered, tumor-derived cell population found in the peripheral blood of cancer patients and are thought to contribute to tumor metastasis. However, identifying CHCs by immunofluorescence (IF) imaging of patient peripheral blood mononuclear cells (PBMCs) is a time-consuming and subjective process that currently relies on manual annotation by laboratory technicians. Additionally, while IF is relatively easy to apply to tissue sections, its application to PBMC smears presents challenges due to the presence of biological and technical artifacts. To address these challenges, we present a robust image analysis pipeline to automate the detection and analysis of CHCs in IF images. The pipeline incorporates quality control to optimize specimen preparation protocols and remove unwanted artifacts, leverages a β-variational autoencoder (VAE) to learn meaningful latent representations of single-cell images, and employs a support vector machine (SVM) classifier to achieve human-level CHC detection. We created a rigorously labeled IF CHC data set including nine patients and two disease sites with the assistance of 10 annotators to evaluate the pipeline. We examined annotator variation and bias in CHC detection and provided guidelines to optimize the accuracy of CHC annotation. We found that all annotators agreed on CHC identification for only 65% of the cells in the data set and had a tendency to underestimate CHC counts for regions of interest (ROIs) containing relatively large amounts of cells (>50,000) when using the conventional enumeration method. On the other hand, our proposed approach is unbiased to ROI size. The SVM classifier trained on the β-VAE embeddings achieved an F1 score of 0.80, matching the average performance of human annotators. Our pipeline enables researchers to explore the role of CHCs in cancer progression and assess their potential as a clinical biomarker for metastasis. Further, we demonstrate that the pipeline can identify discrete cellular phenotypes among PBMCs, highlighting its utility beyond CHCs.

A Hybrid Intelligence Approach for Circulating Tumor Cell Enumeration in Digital Pathology by Using CNN and Weak Annotations

Counting the number of Circulating Tumor Cells (CTCs) for cancer screenings is currently done by cytopathologists with a heavy time and energy cost. AI, especially deep learning, has shown great potential in medical imaging domains. The aim of this paper is to develop a novel hybrid intelligence approach to automatically enumerate CTCs by combining cytopathologist expertise with the efficiency of deep learning convolutional neural networks (CNNs). This hybrid intelligence approach includes three major components: CNN based CTC detection/localization using weak annotations, CNN based CTC segmentation, and a classifier to ultimately determine CTCs. A support vector machine (SVM) was investigated for classification efficiency. The B-scale transform was also introduced to find the maximum sphericality of a given region. The SVM classifier was implemented to use a three-element vector as its input, including the B-scale (size), texture, and area values from the detection and segmentation results. We collected 466 fluoroscopic images for CTC detection/localization, 473 images for CTC segmentation and another 198 images with 323 CTCs as an independent data set for CTC enumeration. Precision and recall for CTC detection are 0.98 and 0.92, which is comparable with the state-of-the-art results that needed much larger and stricter training data sets. The counting error on an independent testing set was 2-3% and 9% (with/without B-scale) and performs much better than previous thresholding approaches with 30% of counting error rates. Recent publications prove facilitation of other types of research in object localization and segmentation are necessary.

Label-free tumor cells classification using deep learning and high-content imaging

Many studies have shown that cellular morphology can be used to distinguish spiked-in tumor cells in blood sample background. However, most validation experiments included only homogeneous cell lines and inadequately captured the broad morphological heterogeneity of cancer cells. Furthermore, normal, non-blood cells could be erroneously classified as cancer because their morphology differ from blood cells. Here, we constructed a dataset of microscopic images of organoid-derived cancer and normal cell with diverse morphology and developed a proof-of-concept deep learning model that can distinguish cancer cells from normal cells within an unlabeled microscopy image. In total, more than 75,000 organoid-drived cells from 3 cholangiocarcinoma patients were collected. The model achieved an area under the receiver operating characteristics curve (AUROC) of 0.78 and can generalize to cell images from an unseen patient. These resources serve as a foundation for an automated, robust platform for circulating tumor cell detection.

A Lightweight and Robust Framework for Circulating Genetically Abnormal Cells (CACs) Identification Using 4-Color Fluorescence In Situ Hybridization (FISH) Image and Deep Refined Learning

Circulating genetically abnormal cells (CACs) constitute an important biomarker for cancer diagnosis and prognosis. This biomarker offers high safety, low cost, and high repeatability, which can serve as a key reference in clinical diagnosis. These cells are identified by counting fluorescence signals using 4-color fluorescence in situ hybridization (FISH) technology, which has a high level of stability, sensitivity, and specificity. However, there are some challenges in CACs identification, due to the difference in the morphology and intensity of staining signals. In this concern, we developed a deep learning network (FISH-Net) based on 4-color FISH image for CACs identification. Firstly, a lightweight object detection network based on the statistical information of signal size was designed to improve the clinical detection rate. Secondly, the rotated Gaussian heatmap with a covariance matrix was defined to standardize the staining signals with different morphologies. Then, the heatmap refinement model was proposed to solve the fluorescent noise interference of 4-color FISH image. Finally, an online repetitive training strategy was used to improve the model's feature extraction ability for hard samples (i.e., fracture signal, weak signal, and adjacent signals). The results showed that the precision was superior to 96%, and the sensitivity was higher than 98%, for fluorescent signal detection. Additionally, validation was performed using the clinical samples of 853 patients from 10 centers. The sensitivity was 97.18% (CI 96.72-97.64%) for CACs identification. The number of parameters of FISH-Net was 2.24 M, compared to 36.9 M for the popularly used lightweight network (YOLO-V7s). The detection speed was about 800 times greater than that of a pathologist. In summary, the proposed network was lightweight and robust for CACs identification. It could greatly increase the review accuracy, enhance the efficiency of reviewers, and reduce the review turnaround time during CACs identification.

The use of an artificial intelligence algorithm for circulating tumor cell detection in patients with esophageal cancer

Despite recent advances in multidisciplinary treatments of esophageal squamous cell carcinoma (ESCC), patients frequently suffer from distant metastasis after surgery. For numerous types of cancer, circulating tumor cells (CTCs) are considered predictors of distant metastasis, therapeutic response and prognosis. However, as more markers of cytopathological heterogeneity are discovered, the overall detection process for the expression of these markers in CTCs becomes increasingly complex and time consuming. In the present study, the use of a convolutional neural network (CNN)-based artificial intelligence (AI) for CTC detection was assessed using KYSE ESCC cell lines and blood samples from patients with ESCC. The AI algorithm distinguished KYSE cells from peripheral blood-derived mononuclear cells (PBMCs) from healthy volunteers, accompanied with epithelial cell adhesion molecule (EpCAM) and nuclear DAPI staining, with an accuracy of >99.8% when the AI was trained on the same KYSE cell line. In addition, AI trained on KYSE520 distinguished KYSE30 from PBMCs with an accuracy of 99.8%, despite the marked differences in EpCAM expression between the two KYSE cell lines. The average accuracy of distinguishing KYSE cells from PBMCs for the AI and four researchers was 100 and 91.8%, respectively (P=0.011). The average time to complete cell classification for 100 images by the AI and researchers was 0.74 and 630.4 sec, respectively (P=0.012). The average number of EpCAM-positive/DAPI-positive cells detected in blood samples by the AI was 44.5 over 10 patients with ESCC and 2.4 over 5 healthy volunteers (P=0.019). These results indicated that the CNN-based image processing algorithm for CTC detection provides a higher accuracy and shorter analysis time compared to humans, suggesting its applicability for clinical use in patients with ESCC. Moreover, the finding that AI accurately identified even EpCAM-negative KYSEs suggested that the AI algorithm may distinguish CTCs based on as yet unknown features, independent of known marker expression.

Label-free liquid biopsy through the identification of tumor cells by machine learning-powered tomographic phase imaging flow cytometry

Image-based identification of circulating tumor cells in microfluidic cytometry condition is one of the most challenging perspectives in the Liquid Biopsy scenario. Here we show a machine learning-powered tomographic phase imaging flow cytometry system capable to provide high-throughput 3D phase-contrast tomograms of each single cell. In fact, we show that discrimination of tumor cells against white blood cells is potentially achievable with the aid of artificial intelligence in a label-free flow-cyto-tomography method. We propose a hierarchical machine learning decision-maker, working on a set of features calculated from the 3D tomograms of the cells' refractive index. We prove that 3D morphological features are adequately distinctive to identify tumor cells versus the white blood cell background in the first stage and, moreover, in recognizing the tumor type at the second decision step. Proof-of-concept experiments are shown, in which two different tumor cell lines, namely neuroblastoma cancer cells and ovarian cancer cells, are used against monocytes. The reported results allow claiming the identification of tumor cells with a success rate higher than 97% and with an accuracy over 97% in discriminating between the two cancer cell types, thus opening in a near future the route to a new Liquid Biopsy tool for detecting and classifying circulating tumor cells in blood by stain-free method.

Enhancing clinical potential of liquid biopsy through a multi-omic approach: A systematic review

In the last years, liquid biopsy gained increasing clinical relevance for detecting and monitoring several cancer types, being minimally invasive, highly informative and replicable over time. This revolutionary approach can be complementary and may, in the future, replace tissue biopsy, which is still considered the gold standard for cancer diagnosis. “Classical” tissue biopsy is invasive, often cannot provide sufficient bioptic material for advanced screening, and can provide isolated information about disease evolution and heterogeneity. Recent literature highlighted how liquid biopsy is informative of proteomic, genomic, epigenetic, and metabolic alterations. These biomarkers can be detected and investigated using single-omic and, recently, in combination through multi-omic approaches. This review will provide an overview of the most suitable techniques to thoroughly characterize tumor biomarkers and their potential clinical applications, highlighting the importance of an integrated multi-omic, multi-analyte approach. Personalized medical investigations will soon allow patients to receive predictable prognostic evaluations, early disease diagnosis, and subsequent ad hoc treatments.

Attention Mask R-CNN with edge refinement algorithm for identifying circulating genetically abnormal cells

Recent studies have suggested that circulating tumor cells with abnormalities in gene copy numbers in mononuclear cell-enriched peripheral blood samples, such as circulating genetically abnormal cells (CACs), can be used as a non-invasive tool to detect patients with benign pulmonary nodules. These cells are identified through fluorescence signals counting by using 4-color fluorescence in situ hybridization (FISH) technology that exhibits high stability, sensitivity, and specificity. When FISH data are analyzed, the overlapping cells and fluorescence noise is a great challenge for identifying of CACs, thereby seriously affecting the efficiency of clinical diagnosis. To address this problem, in this study, we proposed an end-to-end FISH-based method (CACNET) for CAC identification. CACNET achieved nuclear segmentation and counted 4-color staining signals through improved Mask region-based convolutional neural network (R-CNN), followed by cell category (normal cell, deletion cell, gain cell, or CAC) according to pathological criteria. Firstly, the segmentation accuracy of overlapping nuclei was improved by adding an edge constraint head during training. Then, the interference of fluorescence noise was reduced by fusing non-local module to reconstruct the feature extraction network of Mask R-CNN. We trained and tested the proposed model on a dataset comprising 700 frames with 58,083 nuclei. The Accuracy, Sensitivity, and Specificity (overall performance metric for the algorithm) of CAC identification with CACNET were 94.06%, 92.1%, and 99.8%, respectively. Moreover, the developed method exhibited approximately identification speed of approximately 0.22 s per frames. The results showed that the proposed method outperformed the existing CAC identification methods, making it a promising approach for early screening of lung cancer.

Recent Advances in Methods for Circulating Tumor Cell Detection

Circulating tumor cells (CTCs) are released from primary tumors and transported through the body via blood or lymphatic vessels before settling to form micrometastases under suitable conditions. Accordingly, several studies have identified CTCs as a negative prognostic factor for survival in many types of cancer. CTCs also reflect the current heterogeneity and genetic and biological state of tumors; so, their study can provide valuable insights into tumor progression, cell senescence, and cancer dormancy. Diverse methods with differing specificity, utility, costs, and sensitivity have been developed for isolating and characterizing CTCs. Additionally, novel techniques with the potential to overcome the limitations of existing ones are being developed. This primary literature review describes the current and emerging methods for enriching, detecting, isolating, and characterizing CTCs.

Image-based cell sorting using focused travelling surface acoustic waves

Sorting cells is an essential primary step in many biological and clinical applications such as high-throughput drug screening, cancer research and cell transplantation. Cell sorting based on their mechanical properties has long been considered as a promising label-free biomarker that could revolutionize the isolation of cells from heterogeneous populations. Recent advances in microfluidic image-based cell analysis combined with subsequent label-free sorting by on-chip actuators demonstrated the possibility of sorting cells based on their physical properties. However, the high purity of sorting is achieved at the expense of a sorting rate that lags behind the analysis throughput. Furthermore, stable and reliable system operation is an important feature in enabling the sorting of small cell fractions from a concentrated heterogeneous population. Here, we present a label-free cell sorting method, based on the use of focused travelling surface acoustic wave (FTSAW) in combination with real-time deformability cytometry (RT-DC). We demonstrate the flexibility and applicability of the method by sorting distinct blood cell types, cell lines and particles based on different physical parameters. Finally, we present a new strategy to sort cells based on their mechanical properties. Our system enables the sorting of up to 400 particles per s. Sorting is therefore possible at high cell concentrations (up to 36 million per ml) while retaining high purity (>92%) for cells with diverse sizes and mechanical properties moving in a highly viscous buffer. Sorting of small cell fraction from a heterogeneous population prepared by processing of small sample volume (10 μl) is also possible and here demonstrated by the 667-fold enrichment of white blood cells (WBCs) from raw diluted whole blood in a continuous 10-hour sorting experiment. The real-time analysis of multiple parameters together with the high sensitivity and high-throughput of our method thus enables new biological and therapeutic applications in the future.

Machine Learning-Driven Multiobjective Optimization: An Opportunity of Microfluidic Platforms Applied in Cancer Research

Cancer metastasis is one of the primary reasons for cancer-related fatalities. Despite the achievements of cancer research with microfluidic platforms, understanding the interplay of multiple factors when it comes to cancer cells is still a great challenge. Crosstalk and causality of different factors in pathogenesis are two important areas in need of further research. With the assistance of machine learning, microfluidic platforms can reach a higher level of detection and classification of cancer metastasis. This article reviews the development history of microfluidics used for cancer research and summarizes how the utilization of machine learning benefits cancer studies, particularly in biomarker detection, wherein causality analysis is useful. To optimize microfluidic platforms, researchers are encouraged to use causality analysis when detecting biomarkers, analyzing tumor microenvironments, choosing materials, and designing structures.