Analysing cerebrospinal fluid with explainable deep learning: From diagnostics to insights.
cell detection
cerebrospinal fluid
deep learning
explainable AI
heatmaps
Journal
Neuropathology and applied neurobiology
ISSN: 1365-2990
Titre abrégé: Neuropathol Appl Neurobiol
Pays: England
ID NLM: 7609829
Informations de publication
Date de publication:
02 2023
02 2023
Historique:
revised:
14
11
2022
received:
24
06
2022
accepted:
13
12
2022
pubmed:
16
12
2022
medline:
3
3
2023
entrez:
15
12
2022
Statut:
ppublish
Résumé
Analysis of cerebrospinal fluid (CSF) is essential for diagnostic workup of patients with neurological diseases and includes differential cell typing. The current gold standard is based on microscopic examination by specialised technicians and neuropathologists, which is time-consuming, labour-intensive and subjective. We, therefore, developed an image analysis approach based on expert annotations of 123,181 digitised CSF objects from 78 patients corresponding to 15 clinically relevant categories and trained a multiclass convolutional neural network (CNN). The CNN classified the 15 categories with high accuracy (mean AUC 97.3%). By using explainable artificial intelligence (XAI), we demonstrate that the CNN identified meaningful cellular substructures in CSF cells recapitulating human pattern recognition. Based on the evaluation of 511 cells selected from 12 different CSF samples, we validated the CNN by comparing it with seven board-certified neuropathologists blinded for clinical information. Inter-rater agreement between the CNN and the ground truth was non-inferior (Krippendorff's alpha 0.79) compared with the agreement of seven human raters and the ground truth (mean Krippendorff's alpha 0.72, range 0.56-0.81). The CNN assigned the correct diagnostic label (inflammatory, haemorrhagic or neoplastic) in 10 out of 11 clinical samples, compared with 7-11 out of 11 by human raters. Our approach provides the basis to overcome current limitations in automated cell classification for routine diagnostics and demonstrates how a visual explanation framework can connect machine decision-making with cell properties and thus provide a novel versatile and quantitative method for investigating CSF manifestations of various neurological diseases.
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
e12866Informations de copyright
© 2022 The Authors. Neuropathology and Applied Neurobiology published by John Wiley & Sons Ltd on behalf of British Neuropathological Society.
Références
Rahimi J, Woehrer A. Overview of cerebrospinal fluid cytology. Handb Clin Neurol. 2018;145:563-571. doi:10.1016/B978-0-12-802395-2.00035-3
Wick M, Gross CC, Tumani H, Wildemann B, Stangel M, on behalf of the German Society of CSF Diagnostics and Clinical Neurochemistry, DGLN e.V. Automated analysis of cerebrospinal fluid cells using commercially available blood cell analysis devices-a critical appraisal. Cell. 2021;10(5):1232. doi:10.3390/cells10051232
Müller-Jensen L, Diamandis E, Osterloh A, Ruprecht K, Leithner C. CSF cytology in subacute subarachnoid haemorrhage. Neurology. Published online. 2020.
Nagy K, Skagervik I, Tumani H, et al. Cerebrospinal fluid analyses for the diagnosis of subarachnoid haemorrhage and experience from a Swedish study. What method is preferable when diagnosing a subarachnoid haemorrhage? Clin Chem Lab Med (CCLM). 2013;51(11):2073-2086. doi:10.1515/cclm-2012-0783
Perske C, Nagel I, Nagel H, Strik H. CSF cytology-the ongoing dilemma to distinguish neoplastic and inflammatory lymphocytes. Diagn Cytopathol. 2011;39(8):621-626. doi:10.1002/dc.21510
Prayson RA, Fischler DF. Cerebrospinal fluid cytology: an 11-year experience with 5951 specimens. Arch Pathol Lab Med. 1998;122(1):47-51.
Matek C, Krappe S, Münzenmayer C, Haferlach T, Marr C. Highly accurate differentiation of bone marrow cell morphologies using deep neural networks on a large image data set. Blood, J Am Soc Hematol. 2021;138(20):1917-1927. doi:10.1182/blood.2020010568
Wang Q, Bi S, Sun M, Wang Y, Wang D, Yang S. Deep learning approach to peripheral leukocyte recognition. PLoS ONE. 2019;14(6):e0218808. doi:10.1371/journal.pone.0218808
Yu W, Liu Y, Zhao Y, et al. Deep learning-based classification of cancer cell in leptomeningeal metastasis on cytomorphologic features of cerebrospinal fluid. Front Oncol. 2022;12
Matek C, Schwarz S, Spiekermann K, Marr C. Human-level recognition of blast cells in acute myeloid leukaemia with convolutional neural networks. Nat Mach Intell. 2019;1(11):538-544. doi:10.1038/s42256-019-0101-9
Stenzinger A, Alber M, Allgäuer M, et al. Artificial intelligence and pathology: from principles to practice and future applications in histomorphology and molecular profiling. In: Seminars in Cancer Biology. Elsevier; 2021.
Hoberger M, von Laffert M, Heim D, Klauschen F. Histomorphological and molecular profiling: friends not foes! Morpho-molecular analysis reveals agreement between histological and molecular profiling. Histopathology. 2019;75(5):694-703. doi:10.1111/his.13930
Van der Laak J, Litjens G, Ciompi F. Deep learning in histopathology: the path to the clinic. Nat Med. 2021;27(5):775-784. doi:10.1038/s41591-021-01343-4
Tellez D, Litjens G, Bándi P, et al. Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology. Med Image Anal. 2019;58:101544. doi:10.1016/j.media.2019.101544
Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:14091556. Published online 2014.
Bortsova G, Dubost F, Hogeweg L, Katramados I, de Bruijne M. Semi-supervised medical image segmentation via learning consistency under transformations. In: Shen D, Liu T, Peters TM, et al., eds. Medical Image Computing and Computer Assisted Intervention - MICCAI 2019-22nd International Conference, Shenzhen, China, October 13-17, 2019, Proceedings, Part VI. Vol 11769. Lecture notes in computer science. Springer; 2019:810-818. doi:10.1007/978-3-030-32226-7\_90
Lannelongue L, Grealey J, Inouye M. Green algorithms: quantifying the carbon footprint of computation. Adv Sci. 2021;8(12):2100707. doi:10.1002/advs.202100707
McInnes L, Healy J, Melville J. UMAP: uniform manifold approximation and projection for dimension reduction. arXiv preprint arXiv:180203426. Published online 2018.
Anders CJ, Neumann D, Samek W, Müller KR, Lapuschkin S. Software for dataset-wide XAI: from local explanations to global insights with Zennit, CoRelAy, and ViRelAy. arXiv preprint arXiv:210613200. Published online 2021.
Bach S, Binder A, Montavon G, Klauschen F, Müller KR, Samek W. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE. 2015;10(7):e0130140. doi:10.1371/journal.pone.0130140
Montavon G, Samek W, Müller KR. Methods for interpreting and understanding deep neural networks. Dig Sig Process. 2018;73:1-15. doi:10.1016/j.dsp.2017.10.011
Zapf A, Castell S, Morawietz L, Karch A. Measuring inter-rater reliability for nominal data-which coefficients and confidence intervals are appropriate? BMC Med Res Methodol. 2016;16(1):1-10.
Wei T, Simko V. R package ‘Corrplot’: visualization of a correlation matrix; 2021. https://github.com/taiyun/corrplot
Hand DJ, Till RJ. A simple generalisation of the area under the ROC curve for multiple class classification problems. Mach Learn. 2001;45(2):171-186. doi:10.1023/A:1010920819831
Binder A, Bockmayr M, Hägele M, et al. Morphological and molecular breast cancer profiling through explainable machine learning. Nat Mach Intell. 2021;3(4):355-366. doi:10.1038/s42256-021-00303-4
Capper D, Jones DT, Sill M, et al. DNA methylation-based classification of central nervous system tumours. Nature. 2018;555(7697):469-474. doi:10.1038/nature26000
Jurmeister P, Bockmayr M, Seegerer P, et al. Machine learning analysis of DNA methylation profiles distinguishes primary lung squamous cell carcinomas from head and neck metastases. Sci Transl Med. 2019;11(509). doi:10.1126/scitranslmed.aaw8513
Kimura K, Tabe Y, Ai T, et al. A novel automated image analysis system using deep convolutional neural networks can assist to differentiate MDS and AA. Sci Rep. 2019;9(1):1-9. doi:10.1038/s41598-019-49942-z
Shafique S, Tehsin S. Acute lymphoblastic leukemia detection and classification of its subtypes using pretrained deep convolutional neural networks. Technol Cancer Res Treat. 2018;17. doi:10.1177/1533033818802789
Chamberlain MC. Neoplastic meningitis. Curr Neurol Neurosci Rep. 2008;8(3):249-258. doi:10.1007/s11910-008-0038-6
Gleissner B, Chamberlain MC. Neoplastic meningitis. Lancet Neurol. 2006;5(5):443-452. doi:10.1016/S1474-4422(06)70443-4
Le Rhun E, Devos P, Weller J, et al. Prognostic validation and clinical implications of the EANO ESMO classification of leptomeningeal metastasis from solid tumors. Neuro Oncol. 2021;23(7):1100-1112. doi:10.1093/neuonc/noaa298
Cawley GC, Talbot NL. On over-fitting in model selection and subsequent selection bias in performance evaluation. J Mach Learn Res. 2010;11:2079-2107.
Alber M, Lapuschkin S, Seegerer P, et al. iNNvestigate neural networks! J Mach Learn Res. 2019;20(93):1-8.
Samek W, Montavon G, Lapuschkin S, Anders CJ, Müller KR. Explaining deep neural networks and beyond: a review of methods and applications. Proc IEEE. 2021;109(3):247-278. doi:10.1109/JPROC.2021.3060483
Hägele M, Seegerer P, Lapuschkin S, et al. Resolving challenges in deep learning-based analyses of histopathological images using explanation methods. Sci Rep. 2020;10(1):1-12. doi:10.1038/s41598-020-62724-2
Seegerer P, Binder A, Saitenmacher R, et al. Interpretable deep neural network to predict estrogen receptor status from haematoxylin-eosin images. In: Artificial Intelligence and Machine Learning for Digital Pathology. Springer; 2020:16-37. doi:10.1007/978-3-030-50402-1_2
Border SP, Sarder P. From what to why, the growing need for a focus shift toward explainability of AI in digital pathology. Front Physiol. 2022;12:821217. doi:10.3389/fphys.2021.821217
Xing J, Radkay L, Monaco SE, Roth CG, Pantanowitz L. Cerebrospinal fluid cytology of Lyme neuroborreliosis: a report of 3 cases with literature review. Acta Cytol. 2015;59(4):339-344. doi:10.1159/000439160
Hashimoto N, Fukushima D, Koga R, et al. Multi-scale domain-adversarial multiple-instance CNN for cancer subtype classification with unannotated histopathological images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition; 2020:3852-3861.
Djukic M, Trimmel R, Nagel I, et al. Cerebrospinal fluid abnormalities in meningeosis neoplastica: a retrospective 12-year analysis. Fluids Barriers CNS. 2017;14(1):1-7. doi:10.1186/s12987-017-0057-2
Pajtler KW, Witt H, Sill M, et al. Molecular classification of ependymal tumors across all CNS compartments, histopathological grades, and age groups. Cancer Cell. 2015;27(5):728-743. doi:10.1016/j.ccell.2015.04.002
Sturm D, Orr BA, Toprak UH, et al. New brain tumor entities emerge from molecular classification of CNS-PNETs. Cell. 2016;164(5):1060-1072. doi:10.1016/j.cell.2016.01.015
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: Visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE International Conference on Computer Vision; 2017:618-626.
Hui LY, Binder A. Batchnorm decomposition for deep neural network interpretation. In: International Work-Conference on Artificial Neural Networks. Springer; 2019:280-291.