Add fine-tuned Cellpose-SAM model weights and model card

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+models/finetuned/**/models/* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,5 @@
+.DS_Store
+__pycache__/
+*.py[cod]
+.ipynb_checkpoints/
+models/finetuned/**/figures/

README.md

@@ -0,0 +1,76 @@
+---
+license: mit
+tags:
+  - cellpose
+  - image-segmentation
+  - fluorescence-microscopy
+  - bacteria
+  - confocal
+base_model: cellpose/cellpose-sam
+---
+
+# Cellpose-SAM fine-tuned for bacterial segmentation in mouse colon tissue
+
+Fine-tuned weights for [Cellpose-SAM](https://github.com/MouseLand/cellpose) (`cpsam`) to segment individual bacteria in Zeiss Airyscan confocal images of mouse colon tissue sections.
+
+## Model description
+
+The base model (`cpsam`) was fine-tuned on manually annotated BFP-channel patches from 16-bit TIFF stacks acquired on a Zeiss LSM 900 Airyscan at 100× magnification. Images contain bacteria expressing a BFP constitutive reporter, embedded in colon tissue with fecal autofluorescence.
+
+**Key imaging parameters:**
+- Pixel size: 0.035 µm/px
+- Expected bacterial diameter: ~43 px (~1.5 µm)
+- Input channel: BFP (constitutive, used for segmentation only)
+- Stack size: (4, 6323, 6344) uint16
+
+## Available model runs
+
+| Run | Train patches | Val patches | Cells annotated | AP@0.5 base | AP@0.5 fine-tuned |
+|-----|--------------|-------------|-----------------|-------------|-------------------|
+| `4stacks_5x5_norm` | 75 | 25 | 3,913 | 0.641 | **0.724** |
+| `5stacks` | 35 | 8 | 4,641 | 0.885 | 0.842 |
+| `1stack` | — | — | — | — | — |
+
+`4stacks_5x5_norm` is the recommended model (largest annotated patch set, best AP improvement over base).
+
+## How to use
+
+```python
+from cellpose import models
+import numpy as np
+
+# Load fine-tuned model
+model = models.CellposeModel(
+    pretrained_model="path/to/models/finetuned/4stacks_5x5_norm/models/cpsam_4stacks_5x5_norm",
+    gpu=True
+)
+
+# bfp_norm: float32 [0, 1] normalised BFP channel (2D array)
+masks, flows, styles = model.predict(
+    [bfp_norm],
+    diameter=43,        # round(1.5 / 0.035)
+    channels=[0, 0],
+    normalize=True
+)
+```
+
+**Important:** always normalise the BFP channel before inference (`normalize=True` or manual p1–p99 stretch). Measure fluorescence intensities from the raw uint16 stack, not the normalised image.
+
+## Training details
+
+- Base model: `cpsam` (Cellpose-SAM)
+- Epochs: 100
+- Learning rate: 1e-5
+- Weight decay: 0.1
+- Batch size: 1
+- Framework: Cellpose 4.x (`train.train_seg`)
+- Hardware: NVIDIA GPU (CUDA 12.6)
+
+## Intended use
+
+Segmentation of bacteria (~1.5 µm) in Airyscan confocal images of mouse colon tissue for downstream fluorescence ratio measurements (GFP/BFP osmotic stress, RFP/BFP oxidative stress reporters).
+
+Post-segmentation filters recommended:
+1. Size filter: 8–300 px²
+2. Spectral filter: remove objects where min(GFP, RFP) / BFP > 0.6 (fecal autofluorescence)
+3. SYTOX exclusion: remove objects overlapping > 25% with SYTOX+ host nuclei

models/finetuned/1stack/models/cpsam_finetuned

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9efeeb4ef576bdad6f99c3e32dff571776fb5b2a7bac28b11700817d0ac797c5
+size 1218638243

models/finetuned/4stacks_5x5_norm/annotations.json

@@ -0,0 +1,10 @@
+{
+  "annotated_train": [
+
+    "Scene-01-20260416-C3M2_Tcol_1-A01Export-01_c1-4_stack",
+    "Scene-04-20260416-C3M2_Tcol_3-A01Export-06_c1-4_stack",
+    "Scene-05-20260416-C3M2_Pcol_2-A01Export-07_c1-4_stack",
+    "Scene-06-20260416-C3M2_Pcol_1-A01Export-08_c1-4_stack"
+
+  ]
+}

models/finetuned/4stacks_5x5_norm/losses.json

@@ -0,0 +1,206 @@
+{
+  "train_losses": [
+    0.1470970908179879,
+    0.14280604326010993,
+    0.08668827597769753,
+    0.06820661813214732,
+    0.05847628867505894,
+    0.05624830764524328,
+    0.0894672537742493,
+    0.05079159116256051,
+    0.058009657344470424,
+    0.06252908549077499,
+    0.0721242892034934,
+    0.07878532054174381,
+    0.0711975631431172,
+    0.06201121828848651,
+    0.04969973565467323,
+    0.04221136723586824,
+    0.057416409716825004,
+    0.08007866989258522,
+    0.06456684654345736,
+    0.06983763948878428,
+    0.0506325890251901,
+    0.06917757960191617,
+    0.06743577012229555,
+    0.059784916142816655,
+    0.06237870822864352,
+    0.05623223176332734,
+    0.0574417450351757,
+    0.04434795874248569,
+    0.04859505581621003,
+    0.06306201783377521,
+    0.06179325063324844,
+    0.048228384820007096,
+    0.04332713907555444,
+    0.05172139843625094,
+    0.050803053100729204,
+    0.054533105639275166,
+    0.046261301792304345,
+    0.05348889923344056,
+    0.0751100899282028,
+    0.04439351665894113,
+    0.04061061179895963,
+    0.056366308493985946,
+    0.044939486186000674,
+    0.03928713157404369,
+    0.08139273978051885,
+    0.05066678731314217,
+    0.05081006165884901,
+    0.05350084146717563,
+    0.062070576412855495,
+    0.0531386282115515,
+    0.054863547750962124,
+    0.057854709778330286,
+    0.07938782613035679,
+    0.06576605394288587,
+    0.0548466194991488,
+    0.07464743239455857,
+    0.048097253193554936,
+    0.06723481159977382,
+    0.057492569981550334,
+    0.04713209259993164,
+    0.050328729118336926,
+    0.03485392045695335,
+    0.04680498436879134,
+    0.061053627264336684,
+    0.05057701545001085,
+    0.03443690091274523,
+    0.054374227461618525,
+    0.0506487723296353,
+    0.034474156267242505,
+    0.0714527938311221,
+    0.045786460238256645,
+    0.05448499069723766,
+    0.07655500041815685,
+    0.05475753917834178,
+    0.06604362122142145,
+    0.053723814981834345,
+    0.04298036513248614,
+    0.07201862307760167,
+    0.062164929657010363,
+    0.0716463479482142,
+    0.05063831342005869,
+    0.05164919118610366,
+    0.04755301324485724,
+    0.05350877836657067,
+    0.04738918868262165,
+    0.05157914667952961,
+    0.05688467302922315,
+    0.05879420917675209,
+    0.04353352682024706,
+    0.05980892521076991,
+    0.05784763035577877,
+    0.05719790937699145,
+    0.06058727096785636,
+    0.05236009782839877,
+    0.05439545734358641,
+    0.04226019134143522,
+    0.06617063626094023,
+    0.03543460139151042,
+    0.03907904096946974,
+    0.028261175126341794
+  ],
+  "val_losses": [
+    0.20270178876817227,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.05480176730197854,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.05781135312397964,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.04780416376073845,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.047801945098908616,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.049854809293174186,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.044820896815217566,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.04211683547677239,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.041883666682406326,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.04192465856205672,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.042008217318507374,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0,
+    0.0
+  ]
+}

models/finetuned/4stacks_5x5_norm/losses.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4df625a7e50306874fa201b53d021394370498ea6eea8a7cecfb25d1ef5e63f
+size 2130

models/finetuned/4stacks_5x5_norm/models/cpsam_4stacks_5x5_norm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a02909a208a34629f4e8a6a89e5f357bb75916e3ad5dfa6cf33952f272f7af1
+size 1218645236

models/finetuned/4stacks_5x5_norm/run_info.json

@@ -0,0 +1,27 @@
+{
+  "run_name": "4stacks_5x5_norm",
+  "date": "2026-04-24",
+  "annotated_stacks": [
+    "Scene-01-20260416-C3M2_Tcol_1-A01Export-01_c1-4_stack",
+    "Scene-04-20260416-C3M2_Tcol_3-A01Export-06_c1-4_stack",
+    "Scene-05-20260416-C3M2_Pcol_2-A01Export-07_c1-4_stack",
+    "Scene-06-20260416-C3M2_Pcol_1-A01Export-08_c1-4_stack"
+  ],
+  "train_stacks": [
+    "Scene-01-20260416-C3M2_Tcol_1-A01Export-01_c1-4_stack",
+    "Scene-04-20260416-C3M2_Tcol_3-A01Export-06_c1-4_stack",
+    "Scene-06-20260416-C3M2_Pcol_1-A01Export-08_c1-4_stack"
+  ],
+  "val_stacks": [
+    "Scene-05-20260416-C3M2_Pcol_2-A01Export-07_c1-4_stack"
+  ],
+  "n_train_patches": 75,
+  "n_val_patches": 25,
+  "n_cells_total": 3913,
+  "n_epochs": 100,
+  "ap_base": 0.6406,
+  "ap_finetuned": 0.7243,
+  "model_path": "D:\\cellpose-biosensor\\cellpose-biosensor\\models\\finetuned\\4stacks_5x5_norm\\models\\cpsam_4stacks_5x5_norm",
+  "losses_npz": "D:\\cellpose-biosensor\\cellpose-biosensor\\models\\finetuned\\4stacks_5x5_norm\\losses.npz",
+  "losses_json": "D:\\cellpose-biosensor\\cellpose-biosensor\\models\\finetuned\\4stacks_5x5_norm\\losses.json"
+}

models/finetuned/5stacks/models/cpsam_5stacks

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e880368794b885b471f89c81f6b6272f7bd7af51dce8393131f739710467d594
+size 1218641986

models/finetuned/5stacks/run_info.json

@@ -0,0 +1,27 @@
+{
+  "run_name": "5stacks",
+  "date": "2026-04-21",
+  "annotated_stacks": [
+    "Scene-01-20260416-C3M2_Tcol_1-A01Export-01_c1-4_stack",
+    "Scene-04-20260416-C3M2_Tcol_3-A01Export-06_c1-4_stack",
+    "Scene-05-20260416-C3M2_Pcol_2-A01Export-07_c1-4_stack",
+    "Scene-06-20260416-C3M2_Pcol_1-A01Export-08_c1-4_stack",
+    "Scene-09-20260416-C3M2_Dcol_3-A01Export-11_c1-4_stack"
+  ],
+  "train_stacks": [
+    "Scene-01-20260416-C3M2_Tcol_1-A01Export-01_c1-4_stack",
+    "Scene-04-20260416-C3M2_Tcol_3-A01Export-06_c1-4_stack",
+    "Scene-05-20260416-C3M2_Pcol_2-A01Export-07_c1-4_stack",
+    "Scene-09-20260416-C3M2_Dcol_3-A01Export-11_c1-4_stack"
+  ],
+  "val_stacks": [
+    "Scene-06-20260416-C3M2_Pcol_1-A01Export-08_c1-4_stack"
+  ],
+  "n_train_patches": 35,
+  "n_val_patches": 8,
+  "n_cells_total": 4641,
+  "n_epochs": 100,
+  "ap_base": 0.8845,
+  "ap_finetuned": 0.8417,
+  "model_path": "D:\\cellpose-biosensor\\cellpose-biosensor\\models\\finetuned\\5stacks\\models\\cpsam_5stacks"
+}

models/finetuned/latest_model_path.txt

@@ -0,0 +1 @@
+D:\cellpose-biosensor\cellpose-biosensor\models\finetuned\4stacks_5x5_norm\models\cpsam_4stacks_5x5_norm