A review of image analysis and machine learning techniques for automated cervical cancer screening from pap-smear images

William, Wasswa; Ware, Andrew; Basaza-Ejiri, Annabella Habinka; Obungoloch, Johnes

doi:10.1016/j.cmpb.2018.05.034

Cited by 196 publications

(118 citation statements)

References 35 publications

Supporting

Mentioning

103

Contrasting

Unclassified

Order By: Relevance

“…Fortunately, because the majority of urine specimens are processed with liquid‐based cytology glass slides, similar to those of Pap tests, they are suitable for image acquisition. Automation‐assisted screening of the Pap test was one of the first and most successful clinical applications of machine learning in cytopathology . Since then, newer methods of machine learning have been developed.…”

Section: Introductionmentioning

confidence: 99%

Performance of an artificial intelligence algorithm for reporting urine cytopathology

Sanghvi

Allen

Callenberg

et al. 2019

Cancer Cytopathology

View full text Add to dashboard Cite

Background Unlike Papanicolaou tests, there are no commercially available computer‐assisted automated screening systems for urine specimens. Despite The Paris System for Reporting Urinary Cytology, there still is poor interobserver agreement with urine cytology and many cases in which a definitive diagnosis cannot be made. In the current study, the authors have reported on the development of an image algorithm that applies computational methods to digitized liquid‐based urine cytology slides. Methods A total of 2405 archival ThinPrep glass slides, including voided and instrumented urine cytology cases, were digitized. A deep learning computational pipeline with multiple tiers of convolutional neural network models was developed for processing whole slide images (WSIs) and predicting diagnoses. The algorithm was validated using a separate test data set comprised of consecutive cases encountered in routine clinical practice. Results There were 1.9 million urothelial cells analyzed. An average of 5400 urothelial cells were identified in each WSI. The algorithm achieved an area under the curve of 0.88 (95% CI, 0.83‐0.93). Using the optimal operating point, the algorithm's sensitivity was 79.5% (95% CI, 64.7%‐90.2%) and the specificity was 84.5% (95% CI, 81.6%‐87.1%) for high‐grade urothelial carcinoma. Conclusions The authors successfully developed a computational algorithm capable of accurately analyzing WSIs of urine cytology cases. Compared with prior studies, this effort used a much larger data set, exploited whole slide–level and not just cell‐level features, and used a cell gallery to display the algorithm's output for easy end‐user review. This algorithm provides computer‐assisted interpretation of urine cytology cases, akin to the machine learning technology currently used for automated Papanicolaou test screening.

show abstract

Section: Introductionmentioning

confidence: 99%

Performance of an artificial intelligence algorithm for reporting urine cytopathology

Sanghvi

Allen

Callenberg

et al. 2019

Cancer Cytopathology

View full text Add to dashboard Cite

show abstract

“…Bei der Analyse des Nutzens des ML muss man sich immer die Frage stellen, welchen Gewinn der Einsatz von ML im speziellen Umfeld bringt. So erlaubt die automatische Klassifikation von Bildern des Cervix uteri in medizinisch schlecht versorgten Gebieten eine Verbesserung der medizinischen Versorgung, die mit normalen Mitteln nicht zu erreichen wäre [16]. Die oben erwähnte Genauigkeit von 83 % in der Frakturanalyse [10] kann in nicht spezialisierten Zentren schon deutlich über der Genauigkeit der örtlichen Kollegen liegen.…”

Section: In Hand-und Plastischer Chirurgieunclassified

Künstliche Intelligenz und Maschinelles Lernen

Hahn

2019

Handchir Mikrochir plast Chir

View full text Add to dashboard Cite

ZusammenfassungMaschinelles Lernen (ML), ein Teilbereich der Künstlichen Intelligenz, ist die Fähigkeit von Computer-Algorithmen, Wissen aus Beispielen zu lernen, ohne dafür explizit programmiert zu sein und dieses Wissen dann auf unbekannte Fälle anzuwenden. Die Anwendung von ML in der Medizin wird in den nächsten Jahren exponentiell zunehmen. Ärzte sollten daher über die prinzipiellen Arten von ML sowie den Prozess des ML grundlegende Kenntnisse haben. Denn nur so ist es möglich, ML optimal zu nutzen und die Grenzen und Probleme des Verfahrens zu erkennen.

show abstract

“…The proposed pre-trained model has also tested on the Herlev database and achieved the highest classification accuracy of 98.76%. [17,18] presents details of recent studies to diagnose cervical cancer.…”

Section: Introductionmentioning

confidence: 99%

Diagnostic Classification of Cervical Cell Images from Pap Smear Slides

Talo¹

2019

acperpro

View full text Add to dashboard Cite

Analysis of pap smear images under a microscope by experts is a laborious and time-consuming task. Computer aided diagnostic (CAD) systems can simplify this tedious process and allow experts to focus on more critical cases. Effective screening and early diagnosis can help to detect precancerous cells and allow early treatment. In this study, we have used a deep learning approach to classify cervical cell images obtained from pap smear slides. The proposed method automatically classifies cervical cell images into five categories without using any pre-processing on raw input images. We have obtained promising results as compared to the previous studies in the literature. The proposed deep learning model can give a second opinion to clinicians in their daily routines and help them to focus on more complex cases.

show abstract

A review of image analysis and machine learning techniques for automated cervical cancer screening from pap-smear images

Cited by 196 publications

References 35 publications

Performance of an artificial intelligence algorithm for reporting urine cytopathology

Performance of an artificial intelligence algorithm for reporting urine cytopathology

Künstliche Intelligenz und Maschinelles Lernen

Diagnostic Classification of Cervical Cell Images from Pap Smear Slides

Contact Info

Product

Resources

About