BIDS

A benchmark and framework for open-world identification of mixed bacterial cultures from optical phase-contrast microscopy.

At a glance

Item	Value
Images	~120,000 1024×1024 images
Source recordings	40 culture sessions (one per combination), in videos.tar.gz
Resolution	1024×1024×3 (JPEG)
Magnification	1000x total (100x oil-immersion objective, NA 1.25)
Species	6 rod-shaped, motile (bs, bt, fj, ka, mx, pf)
Combinations	40 (6 singles, 12 pairs, 15 triples, 6 quadruples, 1 six-species)
Combinations not collected	3 pairs (bs+bt, bt+fj, ka+pf); all 5-species combinations
Splits	random 80/10/10 + leave-combinations-out (LCO), seed 1337
License	CC BY 4.0
Croissant	croissant.json (Croissant 1.0, core + RAI fields)

Species

Code	Latin name	Gram	Motility	Cell length
bs	Bacillus subtilis	+	peritrichous flagella	4–10 µm
bt	Bacillus thermoamylovorans	+	peritrichous flagella	~4 µm
fj	Flavobacterium johnsoniae	−	gliding	5–10 µm
ka	Klebsiella aerogenes	−	peritrichous flagella	1–3 µm (encapsulated)
mx	Myxococcus xanthus	−	gliding	5–10 µm
pf	Pseudomonas fluorescens	−	polar flagella	1.5–3 µm

Directory layout

.
├── croissant.json                 # Croissant 1.0 metadata (core + RAI)
├── LICENSE                        # CC BY 4.0
├── README.md                      # this file
├── splits.json                    # random + LCO split definitions (seed 1337)
├── images.tar.gz                  # ~120,000 1024×1024 images (folder per combo)
└── videos.tar.gz                  # 40 source MP4 videos (for re-extraction)

After extraction, the layout is:

images/
├── bs/                            # singletons: 3,000 images per combo
│   ├── 00001.jpg
│   ├── 00002.jpg
│   └── ...
├── bs_ka/                         # pairs: combo encoded by underscore-joined codes
│   └── ...
├── bs_ka_fj/                      # triples
└── bs_bt_mx_ka_fj_pf/             # six-species mixture

Each image is a 1024×1024 square region (≥40% of the original microscopy image area) with random rotation and translation, resized to 1024×1024. Three images are drawn per source microscopy image to expand the dataset while maintaining session-level train/val/test separation.

The label vector for any combination is the multi-hot indicator over [bs, bt, fj, ka, mx, pf]. For example, bs_ka_fj → [1, 0, 1, 1, 0, 0].

Splits

splits.json contains both protocols.

Random 80/10/10. Image-level split with a fixed seed, intended for in-distribution closed-set characterisation.

Leave-combinations-out (LCO), seed 1337. Holds out nine entire species combinations (one single, two pairs, three triples, two quadruples, and the full six-species combination) under three constraints:

1. Combination disjointness — held-out combinations never appear in train or val.
2. Species coverage — every species appears in at least one trained-on combination, so the protocol tests compositional generalization rather than novel-class detection.
3. Order coverage — the held-out set spans a range of combination orders so performance can be reported as a function of compositional complexity.

The LCO held-out combinations at seed 1337 are: bt, bs_pf, ka_fj, bs_mx_fj, bs_ka_pf, mx_ka_fj, bs_bt_ka_fj, bs_mx_fj_pf, bs_bt_mx_ka_fj_pf.

Wet-lab protocol

Cultures were inoculated from glycerol stocks into nutrient broth (8 g L⁻¹), sterilised by autoclave (121 °C, 15 min), and grown at 30 °C with orbital shaking at 250 rpm for 72–120 hours. Progress was monitored qualitatively by colour change of the medium. Once each culture reached its characteristic growth stage, Petri dishes were prepared from the broth and imaged on an inverted phase-contrast microscope (100x oil-immersion, NA 1.25). Three pairwise combinations (bs+bt, bt+fj, ka+pf) and the five-species combinations were not collected and are absent from the release.

Tasks supported

1. Multi-label species presence detection in mixed cultures.
2. Compositional generalization (LCO protocol).
3. Open-set rejection (leave-one-class-out evaluation; see paper Section 4.3).
4. Novel-class discovery (multi-label compositional NCD; see paper Section 4.3).

Tasks NOT supported

• Proportion / composition estimation. Ground-truth species ratios are unavailable; growth rates differ across species and the wet-lab procedure records only species presence, not relative abundance.
• Clinical pathogen identification. No stained-smear protocol; the dataset must not be used for clinical decisions without independent validation.

Loading the data

A minimal Python loader using only the standard library:

import json, os
from PIL import Image
import numpy as np

# Assume images.tar.gz has been extracted next to splits.json
with open("splits.json") as f:
    splits = json.load(f)

class_names = splits["class_names"]   # ['bs', 'bt', 'fj', 'ka', 'mx', 'pf']

for entry in splits["splits"]["train"]:
    img = np.array(Image.open(entry["path"]).convert("RGB"))   # 1024 x 1024 x 3
    label = np.asarray(entry["label"], dtype=np.int64)         # 6-vector multi-hot
    combo = entry["video"]                                     # combination token, e.g. "bs_ka"
    # ... your pipeline ...

For the LCO protocol, use the helper at baselines.supervised_multilabel_heldout.select_heldout(seed=1337) in the companion code release for an identical apples-to-apples held-out set.

Citation

@misc{bids2026,
  title        = {{BIDS}: A Phase-Contrast Microscopy Benchmark for
                  Open-World Bacterial Identification},
  author       = {Anonymous Authors},
  year         = {2026},
  note         = {Under review at the NeurIPS 2026 Datasets and Benchmarks Track},
  howpublished = {OpenReview: \url{https://openreview.net/PLACEHOLDER}},
}

(Author and institution information will be added on acceptance.)

Contact

During the double-blind review period, please use the OpenReview discussion thread for the corresponding submission. Public contact information will be added on acceptance.