Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach (2024)

Rutgers, J. J., Bánki, T., van der Kamp, A., Waterlander, T. J., Scheijde-Vermeulen, M. A., van den Heuvel-Eibrink, M. M., van der Laak, J. A. W. M., Fiocco, M., Mavinkurve-Groothuis, A. M. C., & de Krijger, R. R. (2021). Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach. Diagnostic Pathology, 16, Artikel 77. https://doi.org/10.1186/s13000-021-01136-w

Rutgers, Jikke J. ; Bánki, Tessa ; van der Kamp, Ananda et al. / Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors : a pilot study for future algorithmic approach. In: Diagnostic Pathology. 2021 ; Vol. 16.

@article{3109ea3ecb714da2a45a1d193b421c92,

title = "Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach",

abstract = "Background: Histopathological classification of Wilms tumors determines treatment regimen. Machine learning has been shown to contribute to histopathological classification in various malignancies but requires large numbers of manually annotated images and thus specific pathological knowledge. This study aimed to assess whether trained, inexperienced observers could contribute to reliable annotation of Wilms tumor components for classification performed by machine learning.Methods: Four inexperienced observers (medical students) were trained in histopathology of normal kidneys and Wilms tumors by an experienced observer (pediatric pathologist). Twenty randomly selected scanned Wilms tumor-slides (from n = 1472 slides) were annotated, and annotations were independently classified by both the inexperienced observers and two experienced pediatric pathologists. Agreement between the six observers and for each tissue element was measured using kappa statistics (κ).Results: Pairwise interobserver agreement between all inexperienced and experienced observers was high (range: 0.845–0.950). The interobserver variability for the different histological elements, including all vital tumor components and therapy-related effects, showed high values for all κ-coefficients (> 0.827).Conclusions: Inexperienced observers can be trained to recognize specific histopathological tumor and tissue elements with high interobserver agreement with experienced observers. Nevertheless, supervision by experienced pathologists remains necessary. Results of this study can be used to facilitate more rapid progress for supervised machine learning-based algorithm development in pediatric pathology and beyond.",

keywords = "AI (artificial intelligence), Classification, Histopathology, Interobserver variability, Machine learning, Wilms tumor",

author = "Rutgers, {Jikke J.} and Tessa B{\'a}nki and {van der Kamp}, Ananda and Waterlander, {Tomas J.} and Scheijde-Vermeulen, {Marijn A.} and {van den Heuvel-Eibrink}, {Marry M.} and {van der Laak}, {Jeroen A.W.M.} and Marta Fiocco and Mavinkurve-Groothuis, {Annelies M.C.} and {de Krijger}, {Ronald R.}",

note = "Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = dec,

doi = "10.1186/s13000-021-01136-w",

language = "English",

volume = "16",

journal = "Diagnostic Pathology",

issn = "1052-9551",

}

Rutgers, JJ, Bánki, T, van der Kamp, A, Waterlander, TJ, Scheijde-Vermeulen, MA, van den Heuvel-Eibrink, MM, van der Laak, JAWM, Fiocco, M, Mavinkurve-Groothuis, AMC & de Krijger, RR 2021, 'Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach', Diagnostic Pathology, vol. 16, 77. https://doi.org/10.1186/s13000-021-01136-w

Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach. / Rutgers, Jikke J.; Bánki, Tessa; van der Kamp, Ananda et al.
In: Diagnostic Pathology, Vol. 16, 77, 12.2021.

Onderzoeksoutput › Academic › peer review

TY - JOUR

T1 - Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors

T2 - a pilot study for future algorithmic approach

AU - Rutgers, Jikke J.

AU - Bánki, Tessa

AU - van der Kamp, Ananda

AU - Waterlander, Tomas J.

AU - Scheijde-Vermeulen, Marijn A.

AU - van den Heuvel-Eibrink, Marry M.

AU - van der Laak, Jeroen A.W.M.

AU - Fiocco, Marta

AU - Mavinkurve-Groothuis, Annelies M.C.

AU - de Krijger, Ronald R.

N1 - Publisher Copyright:© 2021, The Author(s).

PY - 2021/12

Y1 - 2021/12

N2 - Background: Histopathological classification of Wilms tumors determines treatment regimen. Machine learning has been shown to contribute to histopathological classification in various malignancies but requires large numbers of manually annotated images and thus specific pathological knowledge. This study aimed to assess whether trained, inexperienced observers could contribute to reliable annotation of Wilms tumor components for classification performed by machine learning.Methods: Four inexperienced observers (medical students) were trained in histopathology of normal kidneys and Wilms tumors by an experienced observer (pediatric pathologist). Twenty randomly selected scanned Wilms tumor-slides (from n = 1472 slides) were annotated, and annotations were independently classified by both the inexperienced observers and two experienced pediatric pathologists. Agreement between the six observers and for each tissue element was measured using kappa statistics (κ).Results: Pairwise interobserver agreement between all inexperienced and experienced observers was high (range: 0.845–0.950). The interobserver variability for the different histological elements, including all vital tumor components and therapy-related effects, showed high values for all κ-coefficients (> 0.827).Conclusions: Inexperienced observers can be trained to recognize specific histopathological tumor and tissue elements with high interobserver agreement with experienced observers. Nevertheless, supervision by experienced pathologists remains necessary. Results of this study can be used to facilitate more rapid progress for supervised machine learning-based algorithm development in pediatric pathology and beyond.

AB - Background: Histopathological classification of Wilms tumors determines treatment regimen. Machine learning has been shown to contribute to histopathological classification in various malignancies but requires large numbers of manually annotated images and thus specific pathological knowledge. This study aimed to assess whether trained, inexperienced observers could contribute to reliable annotation of Wilms tumor components for classification performed by machine learning.Methods: Four inexperienced observers (medical students) were trained in histopathology of normal kidneys and Wilms tumors by an experienced observer (pediatric pathologist). Twenty randomly selected scanned Wilms tumor-slides (from n = 1472 slides) were annotated, and annotations were independently classified by both the inexperienced observers and two experienced pediatric pathologists. Agreement between the six observers and for each tissue element was measured using kappa statistics (κ).Results: Pairwise interobserver agreement between all inexperienced and experienced observers was high (range: 0.845–0.950). The interobserver variability for the different histological elements, including all vital tumor components and therapy-related effects, showed high values for all κ-coefficients (> 0.827).Conclusions: Inexperienced observers can be trained to recognize specific histopathological tumor and tissue elements with high interobserver agreement with experienced observers. Nevertheless, supervision by experienced pathologists remains necessary. Results of this study can be used to facilitate more rapid progress for supervised machine learning-based algorithm development in pediatric pathology and beyond.

KW - AI (artificial intelligence)

KW - Classification

KW - Histopathology

KW - Interobserver variability

KW - Machine learning

KW - Wilms tumor

U2 - 10.1186/s13000-021-01136-w

DO - 10.1186/s13000-021-01136-w

M3 - Article

C2 - 34419100

AN - SCOPUS:85113211415

SN - 1052-9551

VL - 16

JO - Diagnostic Pathology

JF - Diagnostic Pathology

M1 - 77

ER -

Rutgers JJ, Bánki T, van der Kamp A, Waterlander TJ, Scheijde-Vermeulen MA, van den Heuvel-Eibrink MM et al. Interobserver variability between experienced and inexperienced observers in the histopathological analysis of Wilms tumors: a pilot study for future algorithmic approach. Diagnostic Pathology. 2021 dec.;16:77. doi: 10.1186/s13000-021-01136-w