Advertisement

Journal of Digital Imaging

, Volume 32, Issue 3, pp 521–533 | Cite as

Multi-objective Parameter Auto-tuning for Tissue Image Segmentation Workflows

  • Luis F. R. Taveira
  • Tahsin Kurc
  • Alba C. M. A. Melo
  • Jun Kong
  • Erich Bremer
  • Joel H. Saltz
  • George TeodoroEmail author
Article

Abstract

We propose a software platform that integrates methods and tools for multi-objective parameter auto-tuning in tissue image segmentation workflows. The goal of our work is to provide an approach for improving the accuracy of nucleus/cell segmentation pipelines by tuning their input parameters. The shape, size, and texture features of nuclei in tissue are important biomarkers for disease prognosis, and accurate computation of these features depends on accurate delineation of boundaries of nuclei. Input parameters in many nucleus segmentation workflows affect segmentation accuracy and have to be tuned for optimal performance. This is a time-consuming and computationally expensive process; automating this step facilitates more robust image segmentation workflows and enables more efficient application of image analysis in large image datasets. Our software platform adjusts the parameters of a nuclear segmentation algorithm to maximize the quality of image segmentation results while minimizing the execution time. It implements several optimization methods to search the parameter space efficiently. In addition, the methodology is developed to execute on high-performance computing systems to reduce the execution time of the parameter tuning phase. These capabilities are packaged in a Docker container for easy deployment and can be used through a friendly interface extension in 3D Slicer. Our results using three real-world image segmentation workflows demonstrate that the proposed solution is able to (1) search a small fraction (about 100 points) of the parameter space, which contains billions to trillions of points, and improve the quality of segmentation output by × 1.20, × 1.29, and × 1.29, on average; (2) decrease the execution time of a segmentation workflow by up to 11.79× while improving output quality; and (3) effectively use parallel systems to accelerate parameter tuning and segmentation phases.

Keywords

Digital pathology Computer-assisted image analysis Cell morphology Cancer Parameter auto-tuning 

Notes

Acknowledgments

This work was supported in part by 1U24CA180924-01A1 from the NCI, R01LM011119-01 and R01LM009239 from the NLM, CNPq, and NIH K25CA181503. This research used resources of the XSEDE Science Gateways program under grant TG-ASC130023.

References

  1. 1.
    Muenzel D, Engels H-P, Bruegel M, Kehl V, Rummeny E, Metz S: Intra-and inter-observer variability in measurement of target lesions: implication on response evaluation according to RECIST 1.1. Radiology and oncology 46(1):8–18, 2012PubMedPubMedCentralCrossRefGoogle Scholar
  2. 2.
    Grilley-Olson JE, Hayes DN, Moore DT, Leslie KO, Wilkerson MD, Qaqish BF, Hayward MC, Cabanski CR, Yin X, Socinski MA: Validation of interobserver agreement in lung cancer assessment: hematoxylin-eosin diagnostic reproducibility for non-small cell lung cancer: the 2004 World Health Organization classification and therapeutically relevant subsets. Archives of pathology & laboratory medicine 137(1):32–40, 2012CrossRefGoogle Scholar
  3. 3.
    Warth A, Stenzinger A, von Brünneck A-C, Goeppert B, Cortis J, Petersen I, Hoffmann H, Schnabel PA, Weichert W: Interobserver variability in the application of the novel IASLC/ATS/ERS classification for pulmonary adenocarcinomas. European respiratory journal 40(5):1221–1227, 2012PubMedCrossRefGoogle Scholar
  4. 4.
    Yoon SH, Kim KW, Goo JM, Kim D-W, Hahn S: Observer variability in RECIST-based tumour burden measurements: a meta-analysis. European Journal of Cancer 53:5–15, 2016PubMedCrossRefGoogle Scholar
  5. 5.
    Nakazato Y, Maeshima AM, Ishikawa Y, Yatabe Y, Fukuoka J, Yokose T, Tomita Y, Minami Y, Asamura H, Tachibana K: Interobserver agreement in the nuclear grading of primary pulmonary adenocarcinoma. Journal of Thoracic Oncology 8(6):736–743, 2013PubMedCrossRefGoogle Scholar
  6. 6.
    Bueno-de-Mesquita J, Nuyten D, Wesseling J, van Tinteren H, Linn S, van De Vijver M: The impact of inter-observer variation in pathological assessment of node-negative breast cancer on clinical risk assessment and patient selection for adjuvant systemic treatment. Annals of oncology 21(1):40–47, 2010PubMedCrossRefGoogle Scholar
  7. 7.
    Matasar M, Shi W, Silberstien J, Lin O, Busam K, Teruya-Feldtein J, Filippa D, Zelenetz A, Noy A: Expert second-opinion pathology review of lymphoma in the era of the World Health Organization classification. Annals of oncology:mdr029, 2011Google Scholar
  8. 8.
    Rizzardi AE, Johnson AT, Vogel RI, Pambuccian SE, Henriksen J, Skubitz AP, Metzger GJ, Schmechel SC: Quantitative comparison of immunohistochemical staining measured by digital image analysis versus pathologist visual scoring. Diagnostic pathology 7(1):42, 2012PubMedPubMedCentralCrossRefGoogle Scholar
  9. 9.
    Berney DM, Algaba F, Camparo P, Compérat E, Griffiths D, Kristiansen G, Lopez-Beltran A, Montironi R, Varma M, Egevad L: The reasons behind variation in Gleason grading of prostatic biopsies: areas of agreement and misconception among 266 European pathologists. Histopathology 64(3):405–411, 2014PubMedCrossRefGoogle Scholar
  10. 10.
    Netto GJ, Eisenberger M, Epstein JI, T. T. Investigators: Interobserver variability in histologic evaluation of radical prostatectomy between central and local pathologists: findings of TAX 3501 multinational clinical trial. Urology 77(5):1155–1160, 2011PubMedCrossRefGoogle Scholar
  11. 11.
    Allsbrook WC, Mangold KA, Johnson MH, Lane RB, Lane CG, Epstein JI: Interobserver reproducibility of Gleason grading of prostatic carcinoma: general pathologist. Human pathology 32(1):81–88, 2001PubMedCrossRefGoogle Scholar
  12. 12.
    Sørensen J, Hirsch F, Gazdar A, Olsen J: Interobserver variability in histopathologic subtyping and grading of pulmonary adenocarcinoma. Cancer 71(10):2971–2976, 1993PubMedCrossRefGoogle Scholar
  13. 13.
    Roggli VL, Vollmer RT, Greenberg SD, McGavran MH, Spjut HJ, Yesner R: Lung cancer heterogeneity: a blinded and randomized study of 100 consecutive cases. Human pathology 16(6):569–579, 1985PubMedCrossRefGoogle Scholar
  14. 14.
    Wilkins BS, Erber WN, Bareford D, Buck G, Wheatley K, East CL, Paul B, Harrison CN, Green AR, Campbell PJ: Bone marrow pathology in essential thrombocythemia: interobserver reliability and utility for identifying disease subtypes. Blood 111(1):60–70, 2008PubMedCrossRefGoogle Scholar
  15. 15.
    Kong J, Cooper LA, Wang F, Gao J, Teodoro G, Scarpace L, Mikkelsen T, Schniederjan MJ, Moreno CS, Saltz JH: Machine-based morphologic analysis of glioblastoma using whole-slide pathology images uncovers clinically relevant molecular correlates. PloS one 8(11):e81049, 2013PubMedPubMedCentralCrossRefGoogle Scholar
  16. 16.
    Cooper LAD, Kong J et al.: Integrated morphologic analysis for the identification and characterization of disease subtypes. Journal of the American Medical Informatics Association 19(2):317–323, 2012PubMedPubMedCentralCrossRefGoogle Scholar
  17. 17.
    Sertel O, Kong J, Shimada H, Catalyurek U, Saltz JH, Gurcan MN: Computer-aided prognosis of neuroblastoma on whole-slide images: classification of stromal development. Pattern recognition 42(6):1093–1103, 2009PubMedPubMedCentralCrossRefGoogle Scholar
  18. 18.
    Kothari S, Phan JH, Stokes TH, Wang MD: Pathology imaging informatics for quantitative analysis of whole-slide images. Journal of the American Medical Informatics Association 20(6):1099–1108, 2013PubMedPubMedCentralCrossRefGoogle Scholar
  19. 19.
    Hsu W, Markey MK, Wang MD: Biomedical imaging informatics in the era of precision medicine: progress, challenges, and opportunities. Journal of the American Medical Informatics Association 20(6):1010–1013, 2013PubMedPubMedCentralCrossRefGoogle Scholar
  20. 20.
    Han D, Wang S, Jiang C, Jiang X, Kim H-E, Sun J, Ohno-Machado L: Trends in biomedical informatics: automated topic analysis of JAMIA articles. Journal of the American Medical Informatics Association 22(6):1153–1163, 2015PubMedPubMedCentralCrossRefGoogle Scholar
  21. 21.
    Blom S, Paavolainen L, Bychkov D, Turkki R, Mäki-Teeri P, Hemmes A, Välimäki K, Lundin J, Kallioniemi O, Pellinen T: Systems pathology by multiplexed immunohistochemistry and whole-slide digital image analysis. Scientific Reports 7(1):15580, 2017PubMedPubMedCentralCrossRefGoogle Scholar
  22. 22.
    Yu K-H, Zhang C, Berry GJ, Altman RB, Ré C, Rubin DL, Snyder M: Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. Nature Communications 7, 2016Google Scholar
  23. 23.
    Romo-Bucheli D, Janowczyk A, Romero E, Gilmore H, Madabhushi A: Automated tubule nuclei quantification and correlation with oncotype DX risk categories in ER+ breast cancer whole slide images. SPIE Medical Imaging, 2016Google Scholar
  24. 24.
    Leo P, Lee G, Madabhushi A: Evaluating stability of histomorphometric features across scanner and staining variations: predicting biochemical recurrence from prostate cancer whole slide images, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, 2016Google Scholar
  25. 25.
    Beck AH, Sangoi AR, Leung S, Marinelli RJ, Nielsen TO, van de Vijver MJ, West RB, van de Rijn M, Koller D: Systematic analysis of breast cancer morphology uncovers stromal features associated with survival. Sci Transl Med 3(108):108ra113, Nov 9, 2011PubMedCrossRefGoogle Scholar
  26. 26.
    Chen XS, Wu JY, Huang O, Chen CM, Wu J, Lu JS, Shao ZM, Shen ZZ, Shen KW: Molecular subtype can predict the response and outcome of Chinese locally advanced breast cancer patients treated with preoperative therapy. Oncology reports 23(5):1213–1220, 2010PubMedGoogle Scholar
  27. 27.
    Gurcan MN, Boucheron LE, Can A, Madabhushi A, Rajpoot NM, Yener B: Histopathological image analysis: a review. IEEE Reviews in Biomedical Engineering 2:147–171, 2009PubMedPubMedCentralCrossRefGoogle Scholar
  28. 28.
    Ahrens MB, Orger MB, Robson DN, Li JM, Keller PJ: Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nature Methods 10:413, 2013PubMedCrossRefGoogle Scholar
  29. 29.
    Bankhead P, Loughrey MB, Fernández JA, Dombrowski Y, McArt DG, Dunne PD, McQuaid S, Gray RT, Murray LJ, Coleman HG, James JA, Salto-Tellez M, Hamilton PW: QuPath: open source software for digital pathology image analysis. Scientific Reports 7(1):16878, 2017PubMedPubMedCentralCrossRefGoogle Scholar
  30. 30.
    Torsney-Weir T, Saad A, Moller T, Hege H-C, Weber B, Verbavatz J-M: Tuner: principled parameter finding for image segmentation algorithms using visual response surface exploration. IEEE Trans. on Visualization and Computer Graphics 17(12):1892–1901, 2011PubMedCrossRefGoogle Scholar
  31. 31.
    Held C, Nattkemper T, Palmisano R, Wittenberg T: Approaches to automatic parameter fitting in a microscopy image segmentation pipeline: an exploratory parameter space analysis. Journal of Pathology Informatics 4(2):5–5, 2013CrossRefGoogle Scholar
  32. 32.
    Budinich M, Bourdon J, Larhlimi A, Eveillard D: A multi-objective constraint-based approach for modeling genome-scale microbial ecosystems. PLOS ONE 12(2):e0171744, 2017PubMedPubMedCentralCrossRefGoogle Scholar
  33. 33.
    Coello CA, Lamont GB, Veldhuizen DAV: Evolutionary algorithms for solving multi-objective problems. Berlin: Springer, 2007Google Scholar
  34. 34.
    Jordan H, Thoman P, Durillo JJ, Pellegrini S, Gschwandtner P, Fahringer T, Moritsch H: A multi-objective auto-tuning framework for parallel codes, in International Conference on High Performance Computing. Networking, Storage and Analysis (SC 12):10:1–10:12, 2012Google Scholar
  35. 35.
    Trivedi A, Srinivasan D, Sanyal K, Ghosh A: A survey of multiobjective evolutionary algorithms based on decomposition. IEEE Transactions on Evolutionary Computation 21(3):440–462, 2017Google Scholar
  36. 36.
    Miettinen K, Mäkelä M: On scalarizing functions in multiobjective optimization. OR Spectrum 24(2), 2002Google Scholar
  37. 37.
    Miettinen KM: Nonlinear multiobjective optimization. Dordrecht: Kluwer Academic Publishers, 1998CrossRefGoogle Scholar
  38. 38.
    Figueira JR, Fonseca CM, Halffmann P, Klamroth K, Paquete L, Ruzika S, Schulze B, Stiglmayr M, Willems D: Easy to say they are hard, but hard to see they are easy— towards a categorization of tractable multiobjective combinatorial optimization problems. Journal of Multi-Criteria Decision Analysis 24(1–2):82–98, 2017CrossRefGoogle Scholar
  39. 39.
    Tabatabaee V, Tiwari A, Hollingsworth JK: Parallel parameter tuning for applications with performance variability. Proc. of the ACM/IEEE Conf. on Supercomputing, 2005Google Scholar
  40. 40.
    Sareni B, Krähenbühl L: Fitness sharing and niching methods revisited. IEEE Transactions on Evolutionary Computation 2(3):97–106, 1998CrossRefGoogle Scholar
  41. 41.
    Jones DR: A taxonomy of global optimization methods based on response surfaces. Journal of Global Optimization 21(4):345–383, 2001CrossRefGoogle Scholar
  42. 42.
    Snoek J, Larochelle H, Adams RP: In: Pereira F, Burges CJC, Bottou L, Weinberger KQ Eds. Practical Bayesian optimization of machine learning algorithms, advances in neural information processing systems, Vol. 25. Red Hook: Curran Associates, Inc., 2012, pp. 2951–2959Google Scholar
  43. 43.
    Morris MD: Factorial sampling plans for preliminary computational experiments. Technometrics 33(2):161–174, 1991CrossRefGoogle Scholar
  44. 44.
    Campolongo F, Cariboni J, Saltelli A: An effective screening design for sensitivity analysis of large models. Environmental Modelling & Software 22(10):1509–1518, 2007CrossRefGoogle Scholar
  45. 45.
    Teodoro G, Kurç TM, Taveira LF, Melo AC, Gao Y, Kong J, Saltz JH: Algorithm sensitivity analysis and parameter tuning for tissue image segmentation pipelines. Bioinformatics:btw749, 2017Google Scholar
  46. 46.
    Rios LM, Sahinidis NV: Derivative-free optimization: a review of algorithms and comparison of software implementations. Journal of Global Optimization 56(3):1247–1293, 2013CrossRefGoogle Scholar
  47. 47.
    Kumar S, Hebert M: Discriminative random fields: a discriminative framework for contextual interaction in classification. In: Proc. 9th IEEE International Conference on Computer Vision, 2003, pp. 1150–1157CrossRefGoogle Scholar
  48. 48.
    Szummer M, Kohli P, Hoiem D: Learning CRFs using graph cuts. In: Proceedings of the 10th European Conference on Computer Vision: Part II, 2008, pp. 582–595Google Scholar
  49. 49.
    McIntosh C, Hamarneh G: Is a single energy functional sufficient? Adaptive energy functionals and automatic initialization. In: Lecture notes in computer science, Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2007, pp. 503–510Google Scholar
  50. 50.
    Schultz T, Kindlmann GL: Open-box spectral clustering: applications to medical image analysis. IEEE Trans. Vis. Comput. Graph. 19(12):2100–2108, 2013PubMedCrossRefGoogle Scholar
  51. 51.
    Taha AA, Hanbury A: Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool. BMC Medical Imaging 15:29–29, 2015PubMedPubMedCentralCrossRefGoogle Scholar
  52. 52.
    Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin J-CC, Pujol S, Bauer C, Jennings D, Fennessy FM, Sonka M, Buatti J, Aylward S, Miller JV, Pieper S, Kikinis R: 3D slicer as an image computing platform for the quantitative imaging network. Magnetic Resonance Imaging 30(9):10, 2012CrossRefGoogle Scholar
  53. 53.
    Gao Y, Ratner V, Zhu L, Diprima T, Kurc T, Tannenbaum A, Saltz J: Hierarchical nucleus segmentation in digital pathology images. SPIE Medical Imaging, 2016Google Scholar
  54. 54.
    Gao Y, Aucoin N, Fedorov A, Fillion-Robin J-C: SlicerPathology, 01/15/2018, 2018; http://www.slicer.org/wiki/Documentation/Nightly/Extensions/SlicerPathology
  55. 55.
    Wang R, Purshouse RC, Fleming PJ: Preference-inspired coevolutionary algorithms for many-objective optimization. IEEE Transactions on Evolutionary Computation 17(4):474–494, 2013CrossRefGoogle Scholar
  56. 56.
    Nelder JA, Mead R: A simplex method for function minimization. The Computer Journal 7(4):308–313, 1965CrossRefGoogle Scholar
  57. 57.
    Jones DR: In: Floudas CA, Pardalos PM Eds. Direct global optimization algorithm, Encyclopedia of Optimization. New York: Springer US, 2001, pp. 431–440Google Scholar
  58. 58.
    Kong J, Cooper L, Wang F, Gao J, Teodoro G, Scarpace L, Mikkelsen T, Schniederjan M, Moreno C, Saltz J: A novel paradigm for determining molecular correlates of tumor cell morphology in human glioma whole slide images. NEURO-ONCOLOGY 15:158–159, 2013Google Scholar
  59. 59.
    Teodoro G, Pan T, Kurc T, Kong J, Cooper L, Klasky S, Saltz J: Region templates: data representation and management for high-throughput image analysis. Parallel Computing 40(10):589–610, 2014PubMedPubMedCentralCrossRefGoogle Scholar
  60. 60.
    Kurç TM, Qi X, Wang D, Wang F, Teodoro G, Cooper LAD, Nalisnik M, Li Z-Y, Saltz JH, Foran DJ: Scalable analysis of big pathology image data cohorts using efficient methods and high-performance computing strategies. BMC Bioinformatics 16:399:1–399:21, 2015CrossRefGoogle Scholar
  61. 61.
    Teodoro G, Kurc T, Kong J, Cooper L, Saltz J: Comparative performance analysis of Intel (R) Xeon Phi (TM), GPU, and CPU: a case study from microscopy image analysis. In: 2014 IEEE 28th International Parallel and Distributed Processing Symposium, 2014, pp. 1063–1072Google Scholar
  62. 62.
    Barreiros W, Teodoro G, Kurc T, Kong J, Melo ACMA, Saltz J: Parallel and efficient sensitivity analysis of microscopy image segmentation workflows in hybrid systems. In: 2017 IEEE International Conference on Cluster Computing (CLUSTER), 2017, pp. 25–35CrossRefGoogle Scholar
  63. 63.
    Aji A, Wang F, Vo H, Lee R, Liu Q, Zhang X, Saltz J: Hadoop GIS: a high performance spatial data warehousing system over mapreduce. Proceedings of the VLDB Endowment 6(11):1009–1020, 2013CrossRefGoogle Scholar
  64. 64.
    Beckmann N, Kriegel H, Schneider R, Seeger B: The R*-tree: an efficient and robust access method for points and rectangles. SIGMOD:322–331, 1990Google Scholar
  65. 65.
    Merkel D: Docker: lightweight Linux containers for consistent development and deployment. Linux J. 2014(239):2, 2014Google Scholar

Copyright information

© Society for Imaging Informatics in Medicine 2018

Authors and Affiliations

  1. 1.Department of Computer ScienceUniversity of BrasíliaBrasíliaBrazil
  2. 2.Department of Biomedical InformaticsStony Brook UniversityStony BrookUSA
  3. 3.Scientific Data GroupOak Ridge National LaboratoryOak RidgeUSA
  4. 4.Department of Biomedical InformaticsEmory University School of MedicineAtlantaUSA
  5. 5.Department of Biomedical EngineeringEmory - Georgia Institute of TechnologyAtlantaUSA
  6. 6.Department of Mathematics and StatisticsGeorgia State UniversityAtlantaUSA

Personalised recommendations