Real time decision support system for diagnosis of rare cancers, trained in parallel, on a graphics processing unit

In the present study a new strategy is introduced for designing and developing of an efficient dynamic Decision Support System (DSS) for supporting rare cancers decision making. The proposed DSS operates on a Graphics Processing Unit (GPU) and it is capable of adjusting its design in real time based on user-defined clinical questions in contrast to standard CPU implementations that are limited by processing and memory constrains. The core of the proposed DSS was a Probabilistic Neural Network classifier and was evaluated on 140 rare brain cancer cases, regarding its ability to predict tumors' malignancy, using a panel of 20 morphological and textural features Generalization was estimated using an external 10-fold cross-validation. The proposed GPU-based DSS achieved significantly higher training speed, outperforming the CPU-based system by a factor that ranged from 267 to 288 times. System design was optimized using a combination of 4 textural and morphological features with 78.6% overall accuracy, whereas system generalization was 73.8%±3.2%. By exploiting the inherently parallel architecture of a consumer level GPU, the proposed approach enables real time, optimal design of a DSS for any user-defined clinical question for improving diagnostic assessments, prognostic relevance and concordance rates for rare cancers in clinical practice.

[1]  Rudy J. Lapeer,et al.  An optimised radial basis function algorithm for fast non-rigid registration of medical images , 2010, Comput. Biol. Medicine.

[2]  Jones Pf,et al.  Errors by locums , 1996 .

[3]  Mark R Gilbert,et al.  Treatment of astrocytomas , 2005, Current opinion in neurology.

[4]  F. Abdul-Karim,et al.  Interobserver reproducibility among neuropathologists and surgical pathologists in fibrillary astrocytoma grading , 2000, Journal of the Neurological Sciences.

[6]  Nabil Belacel,et al.  Multicriteria fuzzy assignment method: a useful tool to assist medical diagnosis , 2001, Artif. Intell. Medicine.

[7]  Jie Yang,et al.  Degree prediction of malignancy in brain glioma using support vector machines , 2006, Comput. Biol. Medicine.

[8]  D. Azria,et al.  Facteurs pronostiques et prise en charge des tumeurs rares , 2006 .

[9]  F. Cowie Treatment of rare cancers: gastrointestinal stromal tumours. , 2006, British journal of hospital medicine.

[10]  Dorin Comaniciu,et al.  Decision support system for multiuser remote microscopy in telepathology , 1999, Proceedings 12th IEEE Symposium on Computer-Based Medical Systems (Cat. No.99CB36365).

[11]  Joel H. Saltz,et al.  Stroma classification for neuroblastoma on graphics processors , 2009, Int. J. Data Min. Bioinform..

[12]  Lee Bowman,et al.  Utilization and Cost of Health Care Services Associated with Primary Malignant Brain Tumors in the United States , 2006, Journal of Neuro-Oncology.

[13]  Michel Barlaud,et al.  Fast k nearest neighbor search using GPU , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[14]  Rodney A. Kennedy,et al.  Parallel computation of mutual information on the GPU with application to real-time registration of 3D medical images , 2010, Comput. Methods Programs Biomed..

[15]  T Joannides Rare cancers. , 2001, Clinical oncology (Royal College of Radiologists (Great Britain)).

[16]  D. Troxel,et al.  Diagnostic Errors in Surgical Pathology Uncovered by a Review of Malpractice Claims , 2000, International journal of surgical pathology.

[17]  Charles Duyckaerts,et al.  The 2007 WHO classification of tumors of the central nervous system – what has changed? , 2008, Current opinion in neurology.

[18]  Keechul Jung,et al.  GPU implementation of neural networks , 2004, Pattern Recognit..

[19]  R Nafe,et al.  Histomorphometry of tumour cell nuclei in astrocytomas using shape analysis, densitometry and topometric analysis , 2005, Neuropathology and applied neurobiology.

[20]  A D Ramsay,et al.  Histopathology departments already audit diagnostic errors , 1996, BMJ.

[21]  P. Lachenbruch An almost unbiased method of obtaining confidence intervals for the probability of misclassification in discriminant analysis. , 1967, Biometrics.

[22]  P Ravazoula,et al.  A computer-based diagnostic and prognostic system for assessing urinary bladder tumour grade and predicting cancer recurrence , 2002, Medical informatics and the Internet in medicine.

[23]  J. Disario,et al.  Clarification of published data , 1995 .

[24]  Anthony Dixon,et al.  Rare skin cancers in general practice. , 2007, Australian family physician.

[25]  D K Pearl,et al.  Improving diagnostic accuracy and interobserver concordance in the classification and grading of primary gliomas , 1997, Cancer.

[26]  Frédéric Maire,et al.  Implementation of Kernel Methods on the GPU , 2005, Digital Image Computing: Techniques and Applications (DICTA'05).

[27]  H. J. Mclaughlin,et al.  Learn , 2002 .

[28]  Donald F. Specht,et al.  Probabilistic neural networks , 1990, Neural Networks.

[29]  C. Fletcher,et al.  Diagnostic histopathology of tumors , 2013 .

[30]  Geoffrey J McLachlan,et al.  Selection bias in gene extraction on the basis of microarray gene-expression data , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[31]  M. Kendrick,et al.  Colorectal cancers of rare histologic types compared with adenocarcinomas , 1994, Diseases of the colon and rectum.

[32]  D. Troxel,et al.  Diagnostic Errors in Surgical Pathology Uncovered by a Review of Malpractice Claims , 2000 .

[33]  Dimitris Glotsos,et al.  Automated segmentation of routinely hematoxylin-eosin-stained microscopic images by combining support vector machine clustering and active contour models. , 2004, Analytical and quantitative cytology and histology.

[34]  Klaus Mueller,et al.  IOP PUBLISHING PHYSICS IN MEDICINE AND BIOLOGY , 2007 .

[35]  H Kalimo,et al.  Grading of diffusely infiltrating astrocytomas by quantitative histopathology, cell proliferation and image cytometric DNA analysis , 2000, Neuropathology and applied neurobiology.

[36]  P. Casali,et al.  Do rare cancers deserve specific strategies for cancer research? , 2010, The Lancet. Oncology.

[37]  P H Bartels,et al.  Computer-assisted discrimination of glioblastomas. , 1997, Analytical and quantitative cytology and histology.

[38]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[39]  D Azria,et al.  [Prognostic factors and management of rare cancers]. , 2006, Cancer radiotherapie : journal de la Societe francaise de radiotherapie oncologique.

[40]  C Decaestecker,et al.  Improving Morphology‐Based Malignancy Grading Schemes in Astrocytic Tumors by Means of Computer‐Assisted Techniques , 1998, Brain pathology.

[41]  A. Vernava,et al.  Rare anal canal cancers in the U.S. veteran: patterns of disease and results of treatment. , 1995, The American surgeon.

[42]  Rajasvaran Logeswaran,et al.  Cholangiocarcinoma—An Automated Preliminary Detection System Using MLP , 2009, Journal of Medical Systems.