Endometriosis often takes years to diagnose. This ultrasound simulation innovation could help change that
Updated
March 17, 2026 1:01 AM
A group of women facing backwards. PHOTO: UNSPLASH
Endometriosis affects roughly one in ten women worldwide, yet diagnosing the condition often takes years. In many cases, patients experience symptoms for nearly a decade before receiving a confirmed diagnosis. One reason is that detecting endometriosis through ultrasound requires specialized training and clinicians do not always encounter enough real cases to build that expertise.
To address this gap, medical simulation company Surgical Science has introduced a new ultrasound training module designed specifically for identifying endometriosis. The system allows clinicians to practice scanning techniques in a virtual environment, helping them recognize signs of the disease without relying solely on real-patient cases.
A key feature of the simulator is training on the “sliding sign,” an ultrasound indicator used to detect deep endometriosis. Because the condition can appear differently from patient to patient, mastering this assessment in real clinical settings can be difficult. The simulator allows clinicians to repeat the process across multiple scenarios, improving their ability to identify the condition during routine examinations.
The module also incorporates the International Deep Endometriosis Analysis (IDEA) protocol, which provides a structured method for performing a complete pelvic ultrasound assessment. Additional training cases, region-based scenarios and certification options are included to support standardized learning.
Early training results suggest strong improvements in clinician confidence, including higher skill levels in transvaginal ultrasound and better recognition of deep endometriosis. By expanding access to structured ultrasound training, simulation tools like this could help reduce diagnostic delays and improve care for millions of women living with the condition.
Keep Reading
What Overstory’s vegetation intelligence reveals about wildfire and outage risk.
Updated
January 15, 2026 8:03 PM
Aerial photograph of a green field. PHOTO: UNSPLASH
Managing vegetation around power lines has long been one of the biggest operational challenges for utilities. A single tree growing too close to electrical infrastructure can trigger outages or, in the worst cases, spark fires. With vast service territories, shifting weather patterns and limited visibility into changing landscape conditions, utilities often rely on inspections and broad wildfire-risk maps that provide only partial insight into where the most serious threats actually are.
Overstory, a company specializing in AI-powered vegetation intelligence, addresses this visibility gap with a platform that uses high-resolution satellite imagery and machine-learning models to interpret vegetation conditions in detail.Instead of assessing risk by region, terrain type or outdated maps, the system evaluates conditions tree by tree. This helps utilities identify precisely where hazards exist and which areas demand immediate intervention—critical in regions where small variations in vegetation density, fuel type or moisture levels can influence how quickly a spark might spread.
At the core of this technology is Overstory’s proprietary Fuel Detection Model, designed to identify vegetation most likely to ignite or accelerate wildfire spread. Unlike broad, publicly available fire-risk maps, the model analyzes the specific fuel conditions surrounding electrical infrastructure. By pinpointing exact locations where certain fuel types or densities create elevated risk, utilities can plan targeted wildfire-mitigation work rather than relying on sweeping, resource-heavy maintenance cycles.
This data-driven approach is reshaping how utilities structure vegetation-management programs. Having visibility into where risks are concentrated—and which trees or areas pose the highest threat—allows teams to prioritize work based on measurable evidence. For many utilities, this shift supports more efficient crew deployment, reduces unnecessary trims and builds clearer justification for preventive action. It also offers a path to strengthening grid reliability without expanding operational budgets.
Overstory’s recent US$43 million Series B funding round, led by Blume Equity with support from Energy Impact Partners and existing investors, reflects growing interest in AI tools that translate environmental data into actionable wildfire-prevention intelligence. The investment will support further development of Overstory’s risk models and help expand access to its vegetation-intelligence platform.
Yet the company’s focus remains consistent: giving utilities sharper, real-time visibility into the landscapes they manage. By converting satellite observations into clear and actionable insights, Overstory’s AI system provides a more informed foundation for decisions that impact grid safety and community resilience. In an environment where a single missed hazard can have far-reaching consequences, early and precise detection has become an essential tool for preventing wildfires before they start.