Inter-laboratory comparison of Ki-67 proliferating index detected by visual assessment and automated digital image analysis

A personal computer-based method was compared with standard visual assessment for quantifying colonization of sweet cherry (Prunus avium L.) leaves by powdery mildew (PM) caused by Podosphaera clandestina (Wallr.:Fr.) Lev. Leaf disks from 14 cultivars were rated for PM severity (percentage of leaf area colonized) by three methods: 1) visual assessment; 2) digital image analysis; and 3) digital image analysis after painting PM colonies on the leaf disk. The third technique, in which PM colonies on each leaf disk were observed using a dissecting microscope and subsequently covered with white enamel paint, provided a standard for comparison of the first two methods. A digital image file for each leaf disk was created using a digital flatbed scanner. Image analysis was performed with a commercially available software package, which did not adequately detect slight differences in color between PM and sweet cherry leaf tissue. Consequently, two replicated experiments revealed a low correlation between PM image analysis and painted PM image analysis (r2 = 0.66 and 0.46, P ≤ 0.0001), whereas visual assessment was highly correlated with painted PM image analysis (r2 = 0.88 and 0.95, P ≤ 0.0001). Rank orders of the 14 cultivars differed significantly (P ≤ 0.05) when PM image analysis and painted PM image analysis were compared; however, rankings by visual assessment were not significantly different (P > 0.05) from those by painted PM image analysis. Thus, standard visual assessment is an accurate method for estimating disease severity in a leaf disk resistance assay for sweet cherry PM.

Download Full-text

Automated digital image analysis and manual counting of Ki-67 proliferation index in patients with breast cancer

Egyptian Journal of Pathology ◽

10.1097/01.xej.0000526541.02481 ◽

2017 ◽

Vol 37 (2) ◽

pp. 228-235

Author(s):

Mervat M.F. El-Deftar ◽

Samir S. Amer ◽

Eman M.O. El-Touny ◽

Amany Aboubakr ◽

Heba El-Zawahry ◽

...

Keyword(s):

Breast Cancer ◽

Image Analysis ◽

Digital Image ◽

Digital Image Analysis ◽

Proliferation Index ◽

Ki 67 ◽

Manual Counting

Download Full-text

Comparison of Manual and Digital Assessment of KI-67 in Neuroendocrine Tumors

American Journal of Clinical Pathology ◽

10.1093/ajcp/aqaa161.274 ◽

2020 ◽

Vol 154 (Supplement_1) ◽

pp. S125-S125

Author(s):

B S Raju ◽

M Quinton ◽

L Hassell

Keyword(s):

Image Analysis ◽

Neuroendocrine Tumors ◽

Digital Image ◽

Digital Image Analysis ◽

Hot Spot ◽

World Health ◽

Proliferation Index ◽

Digital Analysis ◽

Ki 67 ◽

Digital Assessment

Abstract Introduction/Objective Proliferative activity is an essential prognostic and treatment indicator for neuroendocrine tumors (NET). Ki-67 proliferation index, if reported by unaided microscopic estimation on hot-spot locations could lead to variability and inconsistencies. This study aims to compare the Ki-67 assessment of NETs by visual estimation versus automated digital image analysis (Roche iCoreo/Virtuoso). Methods 212 patients with Ki-67-graded GI NETs (117 G1; 61 G2; 34 G3) from 2010 to 2019 were reassessed using digital image analysis quantification of hot spot areas of at least 500 cells (average 800 cells). Revised tumor grades were assigned according to the European Neuroendocrine Tumor Society guidelines and the 2010 World Health Organization classification and compared to initially reported grade. Results We found 75% concordance for G1, with 22% of cases upgraded to G2 and 3% of cases upgraded to G3. For G2, there was 70.5% agreement, with 13.1% of cases downgraded to G1 and 16.4% upgraded to G3. For G3, there was 100% agreement, (kappa=0.64, overall). Retrospective review of discordant G3 cases revealed cases with known metastasis, small fragments of tissue, or polyps. Scanning and scoring required approximately 10 minutes per case. Conclusion Our data shows the time/effort difference of visually estimating versus automated digital analysis may lead to significant classification errors in these tumors. Although digital analysis has limitations, including tumor heterogeneity, misidentification of tumor cells, and poor immunostaining which could require manual counting by a pathologist, this rigor should be reinforced and explicitly stated to increase accuracy and reproducibility of grading.

Download Full-text

Automated assessment of Ki‐67 in breast cancer: the utility of digital image analysis using virtual triple staining and whole slide imaging

Histopathology ◽

10.1111/his.14140 ◽

2020 ◽

Vol 77 (3) ◽

pp. 471-480

Author(s):

Akira I Hida ◽

Dzenita Omanovic ◽

Lars Pedersen ◽

Yumi Oshiro ◽

Takashi Ogura ◽

...

Keyword(s):

Breast Cancer ◽

Image Analysis ◽

Digital Image ◽

Digital Image Analysis ◽

Whole Slide Imaging ◽

Automated Assessment ◽

Ki 67 ◽

Triple Staining

Download Full-text

Audit of KI-67 digital image analysis for breast cancer at pathwest QEII

Pathology ◽

10.1016/j.pathol.2016.12.151 ◽

2017 ◽

Vol 49 ◽

pp. S63

Author(s):

Morgan Wang ◽

C. Thomas ◽

C. Robinson ◽

J. Harvey ◽

G. Sterrett ◽

...

Keyword(s):

Breast Cancer ◽

Image Analysis ◽

Digital Image ◽

Digital Image Analysis ◽

Ki 67

Download Full-text

Nondestructive Measurement of Vegetative Cover Using Digital Image Analysis

HortScience ◽

10.21273/hortsci.39.1.55 ◽

2004 ◽

Vol 39 (1) ◽

pp. 55-59 ◽

Cited By ~ 6

Author(s):

Mercy A. Olmstead ◽

Robert Wample ◽

Stephanie Greene ◽

Julie Tarara

Keyword(s):

Image Analysis ◽

Digital Image ◽

Cover Crops ◽

Cover Crop ◽

Digital Image Analysis ◽

Ground Surface ◽

Visual Assessment ◽

Computer Assisted ◽

Percent Cover ◽

Vegetative Cover

Traditionally, vegetative cover has been subjectively assessed by visual assessment. However, visual assessment is thought to overestimate percent vegetative cover. Thus, a repeatable method to objectively quantify percent cover is desirable. In two vineyards near Prosser, Wash., the percentage of ground surface covered by up to 15 different cover crops was assessed both visually and by computer-assisted digital image analysis. Quadrats in the cover crop were photographed digitally and the images analyzed with commercially available software. Areas of green vegetation in each image were identified and measured. Weeds in some images were differentiated from the cover crop by user-defined thresholds. Subjective visual estimates of percent vegetative cover were generally higher than those digitally estimated. Values for the visual estimates ranged from 5% to 70% in 1998 (mean = 52.4%) and 7.5% to 55% in 1999 (mean = 30.7%), compared to digital readings ranging from 0.5% to 24% (mean = 11.1%) and 10.3% to 36.6% cover (mean = 20.1%), respectively. The visual assessments had lower coefficients of variability in 1998 (cv 28.1) than the digital image analysis (cv 52.3), but in 1999, the values for the two techniques were similar (cv 41.2 vs. cv 45.7). Despite initial variations between the two methods, the accuracy of digital image analysis for measuring percentage vegetative cover is superior.

Download Full-text