Automatic detection of circulating tumor cells in darkfield microscopic images of unstained blood using boosting techniques

Ciurte, Anca; Selicean, Cristina; Şoriţău, Olga; Buiga, Rareș

doi:10.1371/journal.pone.0208385

Cited by 20 publications

(14 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Software-based solutions can be grouped into machine learning image analysis and heuristic-based classification (summarized in Supplementary Table 7). Examples of machine learning image analysis and cell classification solutions include ACCEPT (31,32) and other Convolutional Neural Networks (CNN) (33,34,35). Heuristic-based classification conducts classification based on biophysical properties of cell segments (36).…”

Section: Discussionmentioning

confidence: 99%

Frequent Quantitation of Circulating Tumor Cells Predictive of Real-Time Therapy Response

Lim

Shi

et al. 2022

Preprint

View full text Add to dashboard Cite

Precision medicine is playing an increasingly important role in cancer management and treatment. Specifically in the field of oncology, circulating tumor cells (CTCs) hold significant promise in enabling non-invasive prognostication and near real-time monitoring to individualize treatments. In this study, we present strong associations between CTC subtype counts with treatment response and tumor staging in lung, nasopharyngeal and breast cancers. Longitudinal analysis of CTC count changes over short-time windows further reveals the ability to predict treatment response close to real-time. Our findings demonstrate the suitability of CTCs as a definitive blood-based metric for continuous treatment monitoring. Robust processing of high-throughput image data, explainable classification of CTC subtypes and accurate quantification were achieved using an in-house image analysis system CTC-Quant, which showed excellent agreement with expert opinion upon extensive validation.

show abstract

Section: Discussionmentioning

confidence: 99%

Frequent Quantitation of Circulating Tumor Cells Predictive of Real-Time Therapy Response

Lim

Shi

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Despite the discovery of numerous CTC-specific CSMs, the main limitation that hampers existing CTC detection technologies is still the a priori knowledge of the exact protein composition on the CTCs surfaces, and the lack of a universal marker(s) to address the heterogeneity of CTCs in CRC [125,158,159]. The current gold-standard technique for CTC detection, the microscopic cell imaging, also presents many drawbacks such as the low number of markers, inability to analyze multiple markers simultaneously in routine use, long turnaround time (incompatible with the urgent need for delivery of treatment), and the requirements for specific laboratory instruments and professional expertise (pathologists) for data analysis [160,161]. Furthermore, the lack of large-population follow-up cohort studies increases the difficulties of translating current CSM-based CTC detection methods into the clinical setting for CRC screening, diagnosis, prognosis, real-time monitoring, and therapeutic response [50][51][52][53][54].…”

Section: Challenges In Routine Implementation Of Ctc-specific Csm-dependent Crc Detectionmentioning

confidence: 99%

Liquid Biopsy-Based Colorectal Cancer Screening via Surface Markers of Circulating Tumor Cells

et al. 2021

View full text Add to dashboard Cite

Colorectal cancer (CRC) is ranked second for cancer-related deaths worldwide with approximately half of the patients being diagnosed at the late stages. The untimely detection of CRC results in advancement to the metastatic stage and nearly 90% of cancer-related deaths. The early detection of CRC is crucial to decrease its overall incidence and mortality rates. The recent introduction of circulating tumor cells (CTCs) has enabled a less invasive sampling method from liquid biopsies, besides revealing key information toward CRC metastasis. The current gold standard for CTC identification is the CellSearch® system (Veridex). This first-generation instrumentation relies on a single cell surface marker (CSM) to capture and count CTCs. Detection of CTCs allows the identification of patients at risk for metastasis, whereas CTC enumeration could improve risk assessment, monitoring of systemic therapy, and detection of therapy resistance in advanced metastatic CRC. In this review, we compared the pros and cons between single CSM-based CTC enrichment techniques and multi-marker-based systems. We also highlighted the challenges faced in the routine implementation of CSM-dependent CTC detection methods in CRC screening, prediction, prognosis, disease monitoring, and therapy selection toward precision medicine, as well as the dwelling on post-CTC analysis and characterization methods.

show abstract

“…A similar approach is seen in cancer research, whereby, machine learning has been widely applied in the identification and classification of cancer cells. Similar machine learning models and approaches can be applied in stem cells research [21][22][23][24] that could assist in accelerating the evaluation of safety and efficacy of stem cells. This could potentially bring stem cells to the forefront of personalized medicine.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning in Stem Cells Research: Application for Biosafety and Bioefficacy Assessment

et al. 2021

View full text Add to dashboard Cite

The applicability of machine learning-based analysis in the field of biomedical field has been very beneficial in determining the biological mechanism and validation for a wide range of biological scenarios. This approach is also gaining momentum in variou s stem cells research activities, specifically for stem cells characterization and differentiation pattern. The adoption of similar computational approaches to study and assess biosafety and bioefficacy risks of stem cells for clinical application is the n ext progression. In particular where tumorigenicity has been one of the major concerns in stem cells therapy. There are many factors influencing tumorigenicity in stem cells which may be difficult to capture under conventional laboratory settings. In addition, given the possible multifactorial etiology of tumorigenicity, defining a onesize-fits-all strategy to test such risk in stem cells might not be feasible and may compromise stem cells safety and effectiveness in therapy. Given the increase in biological datasets (which is no longer limited to genomic data) and the advancement of health informatics powered by state-of-the-art machine learning algorithms, there exists a potential for practical application in biosafety and bioefficacy of stem cells therap y. Here, we identified relevant machine learning approaches and suggested protocols intended for stem cells research focusing on the possibility of its usage for stem cells biosafety and bioefficacy assessment. Ultimately, generating models that may assist healthcare professionals to make a better-informed decision in stem cell therapy.

show abstract

Automatic detection of circulating tumor cells in darkfield microscopic images of unstained blood using boosting techniques

Cited by 20 publications

References 18 publications

Frequent Quantitation of Circulating Tumor Cells Predictive of Real-Time Therapy Response

Frequent Quantitation of Circulating Tumor Cells Predictive of Real-Time Therapy Response

Liquid Biopsy-Based Colorectal Cancer Screening via Surface Markers of Circulating Tumor Cells

Machine Learning in Stem Cells Research: Application for Biosafety and Bioefficacy Assessment

Contact Info

Product

Resources

About