
We spoke with the healthtech development company to learn about the real-world AI solutions they've created for healthcare organizations. Thanks, Blackthorn AI, for expert advice and insight on their projects.
Cardiovascular Health & AI
Cardiovascular health depends on accurate data interpretation. Blackthorn partnered with a healthcare startup to develop a prototype system for calculating key cardiovascular metrics like PEP, LVET, and IVCT. The challenge? The startup's wearable sensors provided unstable, noisy data with missing fiducial points.
Blackthorn’s team used a mix of analytical methods and AI tools, including autoencoders and neural networks, to clean and process the data. Within weeks, they built a prototype capable of handling complex signals, delivering validated results that helped test the startup’s business hypothesis.
Breast Cancer Detection
Breast cancer detection often relies on a radiologist’s interpretation of mammograms and ultrasounds. However, diagnoses can have up to 30% error rates. Blackthorn tackled this by developing an AI-powered imaging platform, MedAI, to support radiologists.
By incorporating Vision AI, MedAI analyzed mammography and ultrasound images to spot abnormalities. Clinical trials showed a 20–30% improvement in detection accuracy when the platform was used alongside radiologists. This collaborative approach not only reduced errors but also eased patient anxiety about misdiagnoses.
Fighting the Opioid Crisis with AI
The opioid epidemic in pharma remains a critical issue. The solution by Blackthorn flagged 367 cases of potential overprescription across nearly 12,200 patients in 500 care settings. Each flagged case underwent further scrutiny, preventing misuse and contributing to safer prescribing practices.
How to Reduce Readmission Rates
Hospital readmissions are costly. Blackthorn worked with a provider to address this issue using predictive AI. Analyzing six years of patient data, the system identified readmission risks and categorized patients into three risk groups. Nurses received real-time updates about at-risk patients, allowing them to act promptly.
During a 90-day trial, the AI-driven approach reduced readmissions by 32% and saved the provider $8 million.
Automated Data Analysis for Clinical Trials
Blackthorn supported a drug manufacturer in modernizing data processes for clinical trials. Their goal was to meet CDISC standards while streamlining statistical programming. The team developed workflows for SDTM and ADAM datasets, enabling precise data transformations and statistical analysis.
This work led to successful phase 2 trials of an anti-skin disease drug. By integrating advanced statistical programming tools, Blackthorn made the complex trial process smoother and more transparent, supporting data-driven decision-making.
Personalized Drug Dosage for Rare Diseases
Drug toxicity and resistance pose significant challenges in treating rare diseases. Blackthorn collaborated with a healthcare provider to analyze years of treatment data. Their AI system stratified patients based on susceptibility, identified biochemical factors behind drug resistance, and optimized dosages.
Within 30 days of clinical testing, the system prevented 12 cases of drug poisoning and flagged 3 instances of resistance. Adjusting dosages saved medication costs and improved patient safety. This project demonstrated the potential of AI in personalizing treatments for better outcomes.
For a health tech startup, Blackthorn developed a comprehensive system combining mobile applications, wearable sensors, and VR. The system provided users with health insights, diet advice, and exercise recommendations through an LLM-powered voice assistant. A 3D digital twin and VR studio further enhanced the experience.
This innovative approach to preventive healthcare highlighted the power of combining AI and immersive technologies to improve patient engagement.
Share this post
Leave a comment
All comments are moderated. Spammy and bot submitted comments are deleted. Please submit the comments that are helpful to others, and we'll approve your comments. A comment that includes outbound link will only be approved if the content is relevant to the topic, and has some value to our readers.
Comments (0)
No comment