Skip to main content
Lab Grimoire
TW EN
Coffee
Tools / Image analysis / IHC Batch Analysis
Image analysis

IHC Batch Analysis

Free online batch IHC DAB quantification. Analyze multiple images, compare groups, generate statistics. No upload, runs in browser.

Quantify a batch of IHC images

Upload and check groups

Sample image: CMMI, via scikit-image

Drop or click to add JPEG / PNG / TIFF files · Multiple files supported · 20MB max each (TIFF 50MB)
Groups are inferred from filename prefixes; subject IDs can be inferred from tokens such as S12, mouse4, or patient2, and can be edited.
Images loaded
0
Groups detected
0
Statistical test
At least two groups are needed before a test is run

Subjects are the unit of analysis; multiple fields from one subject are averaged first.

Separation method
DAB threshold: 0.20
Fixed grade cutoffs (OD) (Weak upper bound (OD) / Moderate upper bound (OD))

Fixed OD cutoffs keep the same grading rule across images and are better for batch comparisons.

Guide and FAQ

What batch mode does today

Batch mode applies one shared parameter set to a folder-sized group of cropped hematoxylin + DAB images. Each image is processed in the browser, then the tool summarizes per-image metrics and offers simple between-group statistics on H-Score.

  • Filename-prefilled grouping when your files use a clear prefix such as Control_A1.jpg or Treatment_B2.png.
  • Shared threshold and fixed OD cutoffs so the whole batch is graded with the same rule.
  • Per-image table including positive tissue percentage, mean DAB OD, H-Score, and failure notes.
  • Group comparison on H-Score: a t-test for two groups, or auto-selected ANOVA / Welch / Kruskal-Wallis for three or more groups, with post-hoc comparisons when applicable.
  • Effect-size summary + high-resolution charts: export CSV plus 2240 × 1120 PNG charts for slides or manuscript drafts.

Suggested batch workflow

  1. Name files with a stable group prefix. For example, Control_01.jpg and Treated_01.jpg.
  2. Check the prefilled groups before running. Group labels remain editable in the table.
  3. Keep fixed OD cutoffs consistent. That makes the batch comparable across all uploaded images.
  4. Inspect failed rows. Any file that cannot be analyzed is excluded from the summary, so you should review those before reporting the cohort.
  5. Re-run after changing settings. The UI now flags stale results when the grouping or analysis parameters no longer match the displayed summary.

Current limits

  • The batch assumes one stain protocol. Mixing different scanners, exposures, or stain chemistry in the same batch will reduce comparability.
  • Just like the single-image tool, this is pixel-based. It does not perform cell segmentation or nuclei counting.

Frequently Asked Questions

How many files can I analyze at once?

The limit depends on browser memory and image size. For responsive use, start with cropped ROIs and moderate batch sizes rather than full-slide exports.

What is the box plot showing?

The box plot summarizes positive tissue percentage for each group. Statistical tests, however, are currently performed on H-Score.

Why use fixed OD cutoffs in batch mode?

Relative cutoffs adapt to each image independently, which is useful for exploration. Fixed cutoffs are better when you want a stable grading rule across a cohort.

Can I use this instead of a pathology platform?

Use it as a research-facing quantification helper, not as a clinical diagnostic system. It is best suited to exploratory or internal analysis workflows.

Version history

Version history for all tools is listed in the Chronicle.

View tool updates