AI Templates

AI Template for Cleanliness Analysis of Dental Implants

The mmri.berlin has developed one of the largest and most precise AI models for surface analysis of dental implants. This model has been trained with data from over 150 manufacturers. It detects organic and inorganic contaminations and creates an overview that is essential for quality assurance.

An electron microscope is required: An electron microscope uses an electron beam instead of light to make extremely small structures visible in high resolution.

AI Template for Predicting Surface Functionalization

In collaboration with Dr. Heinrich Jehle, we have created an AI template that predicts specific surface functions of materials.

The created template analyzes surfaces of injection-molded polymer products and calculates the effectiveness of the special surface function: “Cell-repellent effect.”

The template can be used to evaluate products with cell repulsion. It can be employed in the production line to accelerate the setup of injection molding machines by a factor of 1440x.

The template can further be used for controlling similar surface functionalizations, such as:

• Bioinert: Materials that do not cause a reaction with biological tissue, making them particularly suitable for implants.
• Antibacterial: Materials that inhibit or kill the growth of bacteria.
• Biotolerant: Materials that are tolerated by the body without causing a strong immune response.
• Bioactive: Materials that interact positively with tissue, e.g., to promote ingrowth into bone or tissue.

An interferometer is required: A measuring instrument that uses light waves for precise measurement of surfaces or structures to make minimal height differences visible.

AI Template for Universal Object Detection

Template Name: “General Object Detection”

Purpose: This template is universally applicable and designed for the detection of general objects in various environments. It can be used in surveillance systems, automation systems, and for processing images/videos in many areas.

Description:

• Input: Image or video
• Detected Objects: People, vehicles, animals, everyday objects (e.g., bottles, bags, furniture).
• Application: Ideal for general surveillance tasks or scenarios where a variety of objects need to be identified.

Output: Bounding boxes around detected objects with class labels and confidence probabilities.

AI Template for Vehicle Detection and Traffic Analysis

Template Name: “Vehicle Detection & Traffic Analysis”

Purpose: The template is excellently suited for use in traffic monitoring, as it is specifically optimized for various object detection tasks and can be seamlessly integrated into the ProMetronics AI platform. It is frequently used to detect traffic objects such as vehicles, pedestrians, and bicycles.

Description:

• Input: Real-time video feeds or street recordings.
• Detected Objects: Cars, trucks, buses, motorcycles, bicycles, pedestrians.
• Application: Traffic management, traffic counting, detection of traffic violations, smart city solutions.

Output: Bounding boxes around vehicles, pedestrians or bicycles as well as traffic counts and heat maps for traffic density.

AI Template for Analysis of Customer Behavior

Template Name: “Retail Monitoring & Customer Behavior”

Purpose: This template is a powerful model used for detecting human poses and movements. It can be used to measure customer behavior, analyze dwell time, or detect waiting areas and queues.

Description:

• Input: Surveillance cameras
• Detected Objects: People, poses, body parts.
• Application: Analysis of customer movements, dwell time in specific areas, detection of crowded areas or waiting zones.

Output: Heat maps of customer movements, detection of behavioral patterns, counting the number of customers in the area.

AI Template for Anomaly Detection in Medical Images from CT or MRI Scans

Template Name: “Medical Image Object Detection”

Purpose: A model for medical image processing and segmentation. It is optimized for the detection of tumors, organs, and other anomalies in medical images such as CT or MRI scans.

Description:

• Input: Medical image data (X-ray, MRI, CT, ultrasound).
• Detected Objects: Tumors, lesions, bone fractures, organ anomalies.
• Application: Support in diagnosis through segmentation and detection of medical anomalies.

Output: Segmented areas in the images highlighting anomalies or potentially dangerous regions, diagnostic indicators, and confidence probabilities.

Contact