Are you struggling with the challenges of machine learning projects?
- Complex and iterative projects
- Uncertain outcomes
- Rapidly advancing field
- Elusive ML talent
- Lack of ML expertise in pathology
Using the power of machine learning, we help pathology startups turn images into insights:
Detect cancer in novel imaging modalities
Infer molecular biomarkers from H&E histology
Integrate data from different modalities
Find intratumoral heterogeneity
Predict patient outcomes using histology and genomic data
Create robust, generalizable models
Meet Heather D. Couture, PhD
Consultant & Researcher
I leverage my expertise to unlock the value in your data, create more accurate models, and generate new powerful insights from pathology images.
The Power of Machine Learning
- Count mitoses
- Segment glands
- Characterize nuclei shape
- Find tumor in whole slide image
- Predict one imaging modality or stain from another
- Distinguish classes too complex for human experts
- Infer molecular biomarkers from H&E
- Predict patient outcome
Organizations We've Worked With
Featured Pathology Projects
Get the Most Out of Your Images
- Define clear goals and metrics of success
- Properly preprocess data for effective model training
- Iterate quickly to build momentum
- Get from a sufficient to an ideal solution to maximize impact
Ensure the Success of Your Project
Like you, I care about driving impact. I can help you navigate this confusing AI journey. With 15 years of computer vision and machine learning experience, I’ve seen models that fail for a particular task and those that succeed. I have many tools in my toolbox and the research experience to create novel algorithms for unique situations.
My specialty is problems for which there is no existing packaged solution. I make use of today's most powerful machine learning tools - TensorFlow, Keras, PyTorch, sklearn, and others - to help you create a new solution based on images and any other available data.
In collaborating with pathologists, geneticists, epidemiologists, and biostatisticians, I’ve learned a great deal of the common terminology to effectively communicate across domains.
I have been working on pathology applications for more than a decade and have created methods to predict cancer biomarkers too complex for pathologists to see.
Together, we can generate new insights from your project too.
Our team was fortunate to work with Heather on a short-term project, where she was a critical contributor to establishing a framework for future studies. She has an unusual ability to demystify, in plain language, new developments in computer science for a wide audience, and I am confident through her work she will influence many teams and help bring to reality the next set of tools to transform pathology.
Working with Heather and Pixel Scientia Labs was a great choice for us because of her extensive experience in cancer detection, imaging, and machine learning. Based on this background and expertise, she was able to support the team as we worked to empower our imaging solutions through AI and machine learning.
Heather's knowledge of the current state of the art within the digital pathology field is second to none. Discussions of prior art enabled our team to focus on novel research and refine our current AI methodologies for clinical research.
Machine learning projects for pathology have unique challenges
Large & Diverse Images
Whole slide histopathology images are often more than 60,000 pixels across. They contain multiple tissue types and both tumor and non-tumor tissue. Tissue appearance is heterogeneous, both from patient to patient and sometimes within a single tumor.
For some applications like mitosis detection and tissue segmentation, pathologists can provide detailed annotations. But for patient-level prediction tasks like molecular biomarkers, treatment response, or patient outcomes, the algorithm must learn itself which regions of the image are important, often employing multiple instance learning.
Limited Labeled Data
Some applications of computer vision make use of millions of labeled images, whereas data sets of 1000 or so patients are much more common for medical applications. Transfer learning and self-supervised methods are often critical to success in this low sample size regime.
Additional Modalities of Data
The most powerful models that can improve patient care and outcomes often use multiple modalities of data: histopathology, clinical, genomic, proteomic, etc. Specialized models can make use of these structured and unstructured sources of data.
Language Barrier Between Disciplines
Understanding the intricacies of a particular application and possible clinical use cases requires the expertise of pathologists and other domain experts. Project success depends on communicating both ways - about the disease and about the machine learning solution.
The Challenges of Real World Pathology Data
The process for machine learning is empirical and iterative - hypothesize a model, test it, and improve it.
While many talented machine learning engineers can create and train a model, they may be inexperienced with the challenges posed by real world data. The complexities of massive images and noisy or missing labels are a whole different ball game than clean benchmark data sets.
Machine learning engineers may also lack the experience to identify unique aspects of data that, with a customized model, can improve predictions.
I spent my Ph.D. developing solutions to study breast cancer and learned to create new machine learning methodologies motivated by particular aspects of the research data.
The same may be beneficial for your project, but you won’t know it without looking from the right perspective.
Is Deep Learning the Best Approach?
Deep learning is a game changer for many applications. The power of deep learning comes from its ability to find patterns in complex data - even patterns beyond the limits of human perception. It is a new way to gain insights from data. Similar to how we learn from experience, deep learning performs a task repeatedly, tweaking how it does it each time to improve the outcome.
But there are also many situations - like limited training data or interpretability requirements - in which more traditional machine learning might be best. We are well-versed in both methodologies and will help you strategize as your project, your data, and your goals evolve.