EVALUATION OF EFFECTS OF JPEG2000 COMPRESSION ON A COMPUTER-AIDED DETECTION SYSTEM FOR PROSTATE CANCER ON DIGITIZED HISTOPATHOLOGY

被引:11
|
作者
Doyle, Scott [1 ]
Monaco, James [1 ]
Madabhushi, Anant [1 ]
Lindholm, Stefan [2 ]
Ljung, Patric [2 ]
Ladic, Lance [2 ]
Tomaszewski, John [3 ]
Feldman, Michael [3 ]
机构
[1] Rutgers State Univ, Dept Biomed Engn, Piscataway, NJ 08855 USA
[2] Siemens Corp Res, Princeton, NJ USA
[3] Univ Pennsylvania, Dept Surg Pathol, Philadelphia, PA USA
关键词
D O I
10.1109/ISBI.2010.5490238
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
A single digital pathology image can occupy over 10 gigabytes of hard disk space, rendering it difficult to store, analyze, and transmit. Though image compression provides a means of reducing the storage requirement, its effects on CAD (and pathologist) performance are not yet clear. In this work we assess the impact of compression on the ability of a CAD system to detect carcinoma of the prostate (CaP) in histological sections. The CAD algorithm proceeds as follows: Glands in the tissue are segmented using a region-growing algorithm. The size of each gland is then extracted and modeled using a mixture of Gamma distributions. A Markov prior (specifically, a probabilistic pairwise Markov model) is employed to encourage nearby glands to share the same class (i.e. cancerous or non-cancerous). Finally, cancerous glands are aggregated into continuous regions using a distance-hull algorithm. We evaluate CAD performance over 12 images compressed at 14 different compression ratios using JPEG2000. Algorithm performance (measured using the under the receiver operating characteristic curves) remains relatively constant for compression ratios up to 1:256. After this point performance degrades precipitously. We also have an expert pathologist view the compressed images and assign a confidence measure as to their diagnostic fidelity.
引用
收藏
页码:1313 / 1316
页数:4
相关论文
共 50 条
  • [41] Evaluation of an automated computer-aided diagnosis system for the detection of masses on prior mammograms
    Petrick, N
    Chan, HP
    Sahiner, B
    Helvie, MA
    Paquerault, S
    MEDICAL IMAGING 2000: IMAGE PROCESSING, PTS 1 AND 2, 2000, 3979 : 967 - 973
  • [42] Free-response receiver operating characteristic evaluation of lossy JPEG2000 and object-based set partitioning in hierarchical trees compression of digitized mammograms
    Penedo, M
    Souto, M
    Tahoces, PG
    Carreira, JM
    Villalón, J
    Porto, G
    Seoane, C
    Vidal, JJ
    Berbaum, KS
    Chakraborty, DP
    Fajardo, LL
    RADIOLOGY, 2005, 237 (02) : 450 - 457
  • [43] Computer-Aided Detection for Prostate Cancer Detection based on Multi-Parametric Magnetic Resonance Imaging
    Lemaitre, Guillaume
    Marti, Robert
    Rastgoo, Mojdeh
    Meriaudeau, Fabrice
    2017 39TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC), 2017, : 3138 - 3141
  • [44] Development of a computer-aided classification system for cancer detection from digital mammograms
    Alolfe, Mohamed A.
    Youssef, Abo-Bakr M.
    Kadah, Yasser M.
    Mohamed, Ahmed S.
    PROCEEDINGS OF THE 25TH NATIONAL RADIO SCIENCE CONFERENCE: NRSC 2008, 2008,
  • [45] Computer-Aided Diagnosis System for Detection of Stomach Cancer with Image Processing Techniques
    Yasar, Ali
    Saritas, Ismail
    Korkmaz, Huseyin
    JOURNAL OF MEDICAL SYSTEMS, 2019, 43 (04)
  • [46] A computer-aided diagnosis system for breast cancer detection by using a curvelet transform
    Gedik, Nebi
    Atasoy, Ayten
    TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, 2013, 21 (04) : 1002 - 1014
  • [47] Computer-Aided Diagnosis System for Detection of Stomach Cancer with Image Processing Techniques
    Ali Yasar
    Ismail Saritas
    Huseyin Korkmaz
    Journal of Medical Systems, 2019, 43
  • [48] Computer-aided transrectal ultrasound: does prostate HistoScanning™ improve detection performance of prostate cancer in repeat biopsies?
    Hamann, Moritz Franz
    Hamann, C.
    Trettel, A.
    Juenemann, K. P.
    Naumann, C. M.
    BMC UROLOGY, 2015, 15
  • [49] Computer-aided transrectal ultrasound: does prostate HistoScanning™ improve detection performance of prostate cancer in repeat biopsies?
    Moritz Franz Hamann
    C. Hamann
    A. Trettel
    K P Jünemann
    C M Naumann
    BMC Urology, 15
  • [50] Automatic computer-aided detection of prostate cancer based on multiparametric magnetic resonance image analysis
    Vos, P. C.
    Barentsz, J. O.
    Karssemeijer, N.
    Huisman, H. J.
    PHYSICS IN MEDICINE AND BIOLOGY, 2012, 57 (06): : 1527 - 1542