Mednosis LogoMednosis
← All Research

Computer Vision & AI

RSS

Research and developments at the intersection of artificial intelligence and healthcare.

Why it matters: AI is transforming how we diagnose, treat, and prevent disease. Staying informed helps clinicians and patients make better decisions.

Related:DiagnosticsClinical TrialRegulatory

7 research items

Get updates on this topic
Sort by DateSort by Relevance
What Flock’s defenders are missing
From: 2026-08-21
MIT Technology Review - AIExploratory3 min read

Police-Tech Giant Flock Updates Its Massive Vehicle-Tracking Network

Key Takeaway:

Surveillance tech company Flock is updating its 120,000 license plate reader network to curb system misuse, highlighting ongoing debates around algorithmic monitoring and data privacy.

Flock, a major technology company that works with police departments, operates a network of around 120,000 automatic license plate readers across the United States. These automated cameras capture and track vehicle movements. The company recently announced updates to its platform designed to stop misuse and address growing privacy concerns. Understanding how large-scale automated camera systems are managed matters to everyday citizens because automated tracking touches on basic privacy rights. As computer vision tools become common in daily life, establishing clear limits and oversight is essential to keep technological systems accountable and prevent potential abuses.

What this means for you

A major police-technology company is changing how it runs its 120,000 vehicle-tracking cameras to prevent misuse, emphasizing the need for public privacy safeguards as AI monitoring expands.

Computer VisionCardiologyOncology

Citation:

MIT Technology Review - AI, 2026. Read article →

What Flock’s defenders are missing
From: 2026-08-19
MIT Technology Review - AIExploratory2 min read

Flock Updates Nationwide Network of 120,000 License Plate Readers

Key Takeaway:

Surveillance technology provider Flock updated its national network of 120,000 automated license plate recognition cameras to implement new preventative system safeguards.

Flock, a major police-technology company, operates an extensive system of around 120,000 automatic license plate readers across the United States. These cameras constantly scan and log vehicle information in public areas. The company recently announced platform updates designed to prevent certain system issues or misuse. While this technology is used for law enforcement and public safety rather than healthcare, understanding how automated monitoring tools function and how companies attempt to update their safety guardrails is important for anyone interested in modern digital privacy and automated tracking.

What this means for you

Police technology company Flock is updating its national system of 120,000 automated license plate cameras to add new safeguards, though this technology does not affect personal medical care.

Computer VisionCardiologyOncology

Citation:

MIT Technology Review - AI, 2026. Read article →

What Flock’s defenders are missing
From: 2026-08-17
MIT Technology Review - AIExploratory2 min read

Police-Tech Giant Flock Updates 120,000 License Plate Readers

Key Takeaway:

Flock has updated its network of 120,000 automated license plate readers across the United States to implement preventative operational measures.

Flock, a major technology company that works with police departments, operates around 120,000 automated cameras that read vehicle license plates across the United States. The company recently announced new changes to its system designed to prevent specific issues from happening. While the short report does not detail every specific technical fix, it highlights ongoing updates to one of the largest automated camera networks in the country. For everyday citizens, understanding how widespread camera systems are updated helps people stay informed about the technology used in public spaces.

What this means for you

Police technology company Flock updated its system of 120,000 license plate reading cameras. This news relates to public surveillance technology rather than personal healthcare, medical treatments, or clinical advice.

Computer VisionCardiologyOncology

Citation:

MIT Technology Review - AI, 2026. Read article →

Guideline Update
From: 2026-02-18
ArXiv - AI in Healthcare (cs.AI + q-bio)Exploratory3 min read

AI model maps brain tumors to predict patient survival

Key Takeaway:

A new AI model accurately maps brain tumors and predicts survival outcomes, potentially improving treatment planning for glioma patients in neuro-oncology.

Researchers designed an advanced artificial intelligence model to improve the imaging of aggressive brain tumors known as gliomas. The system analyzes magnetic resonance imaging scans using specialized attention mechanisms to map the exact boundaries of the tumor. Beyond mapping, the AI extracts key features from the scans to help predict patient survival outcomes. This dual capability helps neuro-oncologists plan more precise surgeries and customize treatment strategies for individual patients.

What this means for you

This promising research may improve brain tumor treatment in the future, but it's not yet available. Continue following your doctor's advice and don't change your care based on this early study.

Computer VisionOncologyRadiology

Citation:

ArXiv, 2026. arXiv: 2602.15067 Read article →

Guideline Update
From: 2026-02-13
ArXiv - AI in Healthcare (cs.AI + q-bio)Exploratory3 min read

New AI model improves brain tumor detection and survival predictions

Key Takeaway:

A new AI model improves brain tumor detection and survival predictions, potentially aiding precise treatment planning for glioma patients.

Scientists have developed an advanced artificial intelligence model designed to analyze brain scans of patients with gliomas, a highly variable type of brain tumor. The AI uses a specialized deep learning architecture that combines three different viewing angles of the brain to capture detailed spatial context. This allows the model to precisely map the boundaries of the tumor and extract key features that help doctors predict patient survival rates. By providing highly accurate tumor maps and prognosis data, this technology helps neurosurgeons plan safer, more effective surgeries and customize post-operative treatments.

What this means for you

This research is promising but still in early stages. It may take years before it's available. Please continue following your doctor's current recommendations and discuss any concerns with them.

Computer VisionOncologyRadiology

Citation:

ArXiv, 2026. arXiv: 2602.15067 Read article →

Guideline Update
Low-Vision Programmers Can Now Design 3D Models Independently
From: 2026-02-09
IEEE Spectrum - BiomedicalExploratory3 min read

Accessible 3D tools empower low-vision programmers

Key Takeaway:

New 3D modeling tools now allow low-vision programmers to independently create 3D models, improving accessibility in fields like healthcare that require precise design.

Researchers have developed innovative 3D modeling tools that allow low-vision programmers to design complex physical models independently. Historically, 3D design has been highly visual, creating massive barriers for visually impaired individuals in fields like biomedical engineering and prosthetic design. The new system combines physical haptic feedback devices with interactive audio cues, allowing users to feel and hear the shapes they are creating. By removing visual barriers, this tool enables low-vision designers to contribute directly to healthcare innovation and create custom medical devices.

What this means for you

Exciting research for low-vision individuals in 3D modeling, but it's still early. It may take years to become widely available. Continue following your current care plan and consult your doctor for guidance.

OphthalmologyComputer Vision

Citation:

IEEE Spectrum - Biomedical, 2026. Read article →

Guideline Update
Low-Vision Programmers Can Now Design 3D Models Independently
From: 2026-02-06
IEEE Spectrum - BiomedicalExploratory3 min read

New 3D design tools empower low-vision programmers

Key Takeaway:

New 3D modeling tools now allow low-vision programmers to independently create designs, significantly improving accessibility in technology and engineering fields.

Researchers have developed innovative 3D modeling tools designed specifically for low-vision programmers and engineers. Traditional 3D design software relies heavily on visual feedback, creating a massive barrier for visually impaired individuals trying to work in hardware design, robotics, and biomedical engineering. To solve this, the new tools incorporate non-visual interfaces, including auditory feedback and haptic touch technology. This advancement allows low-vision users to independently create complex 3D models, opening up career paths in engineering and ensuring that future medical technologies are designed by a more diverse group of creators.

What this means for you

Exciting early research allows low-vision programmers to design 3D models independently. It's not yet available for public use. Please continue following your current care plan and consult your doctor for guidance.

OphthalmologyComputer Vision

Citation:

IEEE Spectrum - Biomedical, 2026. Read article →

Exploring Computer Vision? Learn how to interpret medical AI research →