Skip to content
Tihum Kabir
All case files

Case file 01 · Computer Vision / Medical AI

ScintNet-KD: Spotting bone metastasis in whole-body scans

Capstone research · North South University (ECE) · Seven authors, faculty-supervised · first author (equal contribution)

  • PyTorch
  • Knowledge Distillation
  • Grad-CAM++
  • LIME
  • Dynamic Quantization
  • Python
  • NumPy

Beat 01

The Dilemma

Whole-body bone scans are the go-to screen for cancer that has spread to bone. Reading them is slow, tiring work, readers don't always agree, and harmless wear-and-tear hotspots are easy to mistake for disease.

Status: Critical · Clinical screening bottleneck

Screening mode
Manual whole-body reads
Reader fatigue
High at volume
Confounders
Harmless wear-and-tear hotspots
Failure mode
Reader disagreement

Baseline: qualitative problem signals

Beat 02

The Hypothesis

If a team of very different CNNs can agree on a diagnosis, one small network should be able to learn their judgment, and small enough to be practical in a busy clinic.

System blueprint

  1. STAGE 01Whole-body scintigraphyAnterior / posterior bone scans
  2. STAGE 0212 CNN teachersEight families, fine-tuned from ImageNet
  3. STAGE 03Best-4 ensemble0.979 F1 · 0.997 AUC, AUC-weighted voting
  4. STAGE 04Knowledge distillationBlended focal loss on teacher + true labels
  5. STAGE 05ScintNet3.27M params, anisotropic kernels, hybrid pooling
  6. STAGE 06Dynamic quantization13.11 MB → 9.96 MB (head only)
  7. STAGE 07Grad-CAM++ · LIME · uptake reportIs it looking at real hotspots?

Beat 03

The Execution

Fine-tuned 12 ImageNet CNNs from eight families, then searched for the four whose errors complement each other: ResNet18, ResNet50, InceptionV3 and EfficientNetB3 (0.979 F1, 0.997 AUC together). Distilled them into ScintNet, a 3.27M-parameter student shaped for tall 1024×256 scans, and quantized its head from 13.11 MB to 9.96 MB.

Knowledge distillation · interactive

Numbers from the paper

Teacher pool

  • ResNet18
  • DenseNet121
  • MobileNetV3-L
  • EfficientNetB3
  • ConvNeXt-T
  • ResNet50
  • EfficientNetB0
  • Xception
  • VGG16
  • EffNetV2-S
  • InceptionV3
  • EfficientNetB1

Student

ScintNet

3.27M params · hybrid pooling head

F1 0.944AUC 0.975Spec 0.990

○ Full-precision weights · 13.11 MB

Twelve ImageNet CNNs from eight families are fine-tuned on whole-body scans, each one a diagnostician in its own right.

Validation

Every positive call comes with Grad-CAM++ heatmaps, LIME regions and an uptake report by body region, so a clinician can check it's looking at real hotspots and not scanner noise. Honest caveat: it still needs external, prospective validation.

ATTENDEDATTENDEDATTENDEDIGNORED

Illustrative synthetic render · not patient data

Explainability audit

Right for the right reasons

A model can score well by latching onto artefacts. Overlaying Grad-CAM and LIME attributions checks that ScintNet's evidence sits on true skeletal lesions, and that degenerative uptake such as arthritic knees is not what drives a positive call.

70%
  • Thoracic spine lesionAttended
  • Left rib lesionAttended
  • Iliac lesionAttended
  • Degenerative knee uptakeIgnored

Beat 04

The Impact

A student 3.3× smaller than its lightest teacher that keeps 97.7% of the ensemble's AUC, and shows its evidence.

Accuracy
94.9%
Held-out test set
F1 Score
0.944
Distilled student
AUC
0.975
Distilled student
Specificity
99.0%
1 false alarm in 99 healthy scans
Sensitivity
87.9%
Caught 51 of 58 metastases
Ensemble AUC
0.997
Best-4 teacher ensemble

ScintNet on the 157-scan held-out test set at the recall-aware threshold (0.570). Ensemble AUC for reference. Needs external validation.

Next case fileBangladesh Legal RAG Chatbot