This week’s list of top data news highlights covers June 15, 2025 to June 22, 2025 and includes articles on detecting melanoma using AI and playing badminton with a robot.
1. Restoring Australia’s Kelp Forests
Google Australia is using AI and satellite data to help regenerate giant kelp forests along the Great Southern Reef in Oceania, a vital but underrecognized marine ecosystem. Through its Digital Future Initiative, the company is working with research institutions to map 7,000 km² of kelp canopy, identify heat-resistant kelp strains, and support targeted replanting efforts. In Tasmania, giant kelp now covers just 5 percent of its historical range—but early results show that AI-driven restoration could reverse this decline.
2. Flagging Abnormal Chest X-Rays
Leeds Teaching Hospitals NHS Trust in the UK is rolling out AI software to help radiologists interpret chest X-rays more efficiently. The tool is trained to detect up to 85 types of clinical findings, ranging from lung cancer and pneumonia to fluid buildup, collapsed lungs, or misplaced breathing tubes, and acts as a diagnostic co-pilot by flagging potential abnormalities for clinicians to review. By speeding up diagnosis and reducing the chance of oversight, the tool aims to get patients the right treatment more quickly and with greater confidence.
Peregrine Technologies, a software company based in California, has developed a platform that emergency responders are using to process complex incidents in real-time. The platform integrates data from sources such as license plate readers, vehicle crash sensors, and 911 call centers, then analyzes it to flag serious events, suggest what resources to dispatch, and support coordination across agencies. Responders access the system through a dashboard already in use at departments around the country.
Johannesburg Water, based in South Africa, is installing smart pressure controllers and leak-detection sensors across its water network to prevent pipe bursts and reduce water loss. The system combines real-time pressure monitoring and acoustic sensors that detect unusual sounds in the pipes, helping teams identify leaks early—especially in high-demand zones and older pipelines. The company plans to have 100 smart controllers installed across its aging water network by the end of this month.
5. Monitoring Cow Health from the Inside
Purdue University researchers have developed a medical robot that monitors cows’ digestive systems from the inside over extended periods. Unlike traditional tools like blood tests, which offer only brief snapshots, this device moves through the cow’s stomach and records internal data—such as methane levels, temperature, and pH—every 10 to 20 seconds throughout the day. Its continuous, real-time tracking gives researchers a richer picture of digestive health and agricultural emissions, helping improve both animal care and environmental monitoring.
MoleMap, a New Zealand-based health-tech company, is using AI to bring expert skin cancer screening to patients in rural and underserved areas across Australia and New Zealand. Specially trained nurses take high-quality images of skin lesions in local clinics and mobile units, and upload these images to a central system where MoleMap’s AI tool, Kāhu AI, flags potential cancers for dermatologists working remotely to review. This model expands access to care while easing demand on specialist resources.
7. Streamlining Art Appraisals
Appraisal Bureau, a New York-based startup, is using AI to speed up professional art valuations for collectors, insurers, and lenders. Its platform analyzes data from more than 20 sources, including gallery sales and market indicators, to estimate values and identify comparables. Certified appraisers employed or contracted by Appraisal Bureau review and sign off on the results to meet legal and professional standards. Investment firms like Oaktree Capital and Hivemind Capital Partners are already using the service to obtain valuations for artworks and NFTs, benefiting from faster turnaround and lower costs than traditional appraisal firms.
8. Predicting Postpartum Depression
Mass General Brigham researchers have developed a machine learning model to identify postpartum depression risk before new mothers leave the hospital. The tool analyzes electronic health records, sociodemographic factors, and prenatal screening data to flag patients most at risk. In tests, the model ruled out postpartum depression in 90 percent of cases and correctly predicted nearly 30 percent of high-risk patients who developed the condition within 6 months.
Rokid, a China-based AR company, has launched Rokid Glasses with built-in payment functionality. The glasses allow users to complete transactions hands-free by scanning QR codes and confirming by voice, with payment details displayed on the lens. The launch marks merging commerce with wearables, reducing transaction time compared to mobile QR payments.
10. Playing Badminton with a Robot
ETH Zurich, a Swiss university, has developed a four-legged robot that can play badminton against humans. The robot uses AI models trained through reinforcement learning to master footwork, track shuttlecock trajectories, and time its swings. Engineers taught the system to mimic real badminton players by analyzing gameplay and continuously adjusting based on trial and error.
