What challenges did American Cancer Society face before AI implementation?

Breast cancer image analysis was slow, subjective, and resource-intensive, requiring manual efforts by trained pathologists. Processing proprietary raw tissue images into a usable format was technically challenging.

What was the impact of the AI solution?

Achieved 12x faster image analysis for breast cancer research versus manual methods. Improved accuracy, consistency, and quality of pathology image interpretations by reducing human fatigue and bias. Established a scalable, reliable cloud platform for future epidemiology imaging research.

Source

GCPEvidence: Medium50/100

American Cancer Society Machine Learning for Breast Cancer Imaging

Use case typeMedical imagingUpdated Jun 13, 2026

The American Cancer Society partnered with Slalom to accelerate and improve the accuracy of breast cancer image analysis for epidemiologic research. Using Google Cloud Cloud ML Engine, Cloud Storage, and TensorFlow, an end-to-end machine learning pipeline was developed for imaging preprocessing, training, and clustering. ML accelerated image analysis by 12 times, improved consistency and objectivity by removing human limitations, and enabled better understanding of breast cancer tissue patterns for future research.

Organization: American Cancer Society
Industry: Healthcare
Location: United States
Published: May 2026

Reported outcomes

12x

timeTime & speed

Strategic outcomes

Speed & agilityAccelerated breast cancer image analysisCustomer experience & trustImproved interpretation accuracy and consistencyScale & capacityEstablished scalable research platformBetter decisions & insightEnabled better tissue pattern understanding

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Time: 12 x increase

Google Cloud Customer StoriesMay 10, 2026Customer storyInferred claimMedium evidence strength

Achieved 12x faster image analysis for breast cancer research versus manual methods.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: American Cancer Society
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Customer Stories

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Medical Imaging Analysis
2Epidemiologic Research
3Machine Learning Pipeline

Breast cancer image analysis was slow, subjective, and resource-intensive, requiring manual efforts by trained pathologists.
Processing proprietary raw tissue images into a usable format was technically challenging.

Implemented an ML pipeline on Google Cloud using Cloud ML Engine for model training and batch prediction.
Stored and processed images with Cloud Storage, converting them into tiled, color-normalized image tiles for ML feature extraction.
Used TensorFlow-based autoencoder models and unsupervised deep learning for pattern recognition and clustering of tissue images.

Technologies

Cloud ML Engine Cloud Storage TensorFlow

Achieved 12x faster image analysis for breast cancer research versus manual methods.
Improved accuracy, consistency, and quality of pathology image interpretations by reducing human fatigue and bias.
Established a scalable, reliable cloud platform for future epidemiology imaging research.

Architecture

End-to-end ML pipeline uses Cloud Storage for image data, Cloud ML Engine for training and inference, and TensorFlow models for autoencoder-based feature extraction and clustering to identify patterns in pathology images.

Implementation partners1

Slalom

Sources & evidence1

Evidence: Medium50/100Evidence strength