Accelerating Chest Radiology with AI in Africa

Learn how our deep learning algorithm labeled 420 chest radiographs in 1.5 minutes - at a level comparable to practicing radiologists

Register & Download

We assist radiologists to reduce their workload with the precision of AI

For radiologists, it can be difficult at times to diagnose with confidence under uncertainty and time pressure. Our AI helps with accurate and fast image diagnosis

  • Beyond expert-level detection of abnormal findings
  • Significant decrease in overall reading time
  • Decrease in unnecessary tests
Read More
116,756

Number of X-Ray Scans collected in Asia, EU and US

349

Number of COVID-19 CT Scans

15

Machine Learning models

4.7

Images processed per second

CheXRad: a trained AI to predict diseases on x-ray images

We developed CheXRad, a deep learning algorithm that identifies locations in the chest radiograph that have predicted probability for 15 different diseases, including COVID-19

CheXRad helps to diagnose Pathologies 160x faster

CheXRad can diagnose certain pathologies in chest radiographs at a level comparable to practicing radiologists on metrics like accuracy, sensitivity and specificity. The big difference in the performance is speed

Features

Developed using Accrad’s cutting-edge deep learning technology

Accrad Train

Trained with a large-scale, high-quality (clinically/CT-proven cases) training set

AI Model Manager

An AI model management tool with the ability to store and manage models locally through user inputs, pull models in from external stores via direct download services to create and manage model catalogs

Performance

Our model performed at an average AUROC 0.9991 across all the epochs. That is, our model's predictions are 99.91% correct on average across all classification thresholds

Accrad Deploy

An extensible reference development framework that facilitates turning AI models into AI-powered clinical workflows with built-in support for DICOM communication and the ability to interface with existing hospital infrastructures

Platform

Our platform is open to third-party developers. You can upload your algorithms and help doctors save lives while generating passive income. We do the heavy-lifting of Clinical Trials and customer acquisition

Monetize your code

We enable developers monetize their algorithms on our platform. You have algorithms which can help our clients? Monetize it here

Open Source

Make your open-source code available at Accrad and help doctors save lives while generating passive income for you

Scalability

Our Platform is built on accelerated deep learning processors for scalable inference workloads in the Cloud

Accrad is building Africa’s largest medical AI software company

We are devoted to conquering cancer and other pathologies across Africa. Our AI software assists radiologists save time and diagnose with accuracy under pressure

We make AI software that can save lives and make it better. We promote creativity. Ideas can come from intuitions, but the final product comes from robust evidence.

This is our journey. You can become part of it

  • 2017: TMIP Brain Tumor Recognition System founded
  • 2018: TMIP Presented at Intel AI Developer Conference
  • 2019: TMIP Chest Radiograph Annotation System
  • 2020: TMIP v2 COVID-19 Recognition System
  • 2020: Renamed TMIP to Accrad (Accelerated Radiology)

Team

We are devoted to conquering cancer and other chest pathologies across Africa by assisting radiologists. Our mission is to deliver 100% accurate diagnosis using AI as the Radialogist's assistant.

Shahram Rezasade

Chief Executive Officer

Co-Founder

Thabo Koee

Chief Technology Officer

Co-Founder

Moloti Tebogo Nakampe

Head of Applied AI

Co-Founder

Amir Rezasade

Chief Medical Officer

Radiation Oncology

Frequently Asked Questions

  • What is an algorithm?

    An algorithm is a set of step-by-step instructions which can be followed to accomplish a goal or solve a problem. A cooking recipe can be an algorithm, just like directions to the hospital. Most of the time, however, we refer to computer algorithms. These are pieces of computer code aimed at solving specific problems. You insert data, the computer algorithm performs calculations based on this data, and gives you an output; the solution to the problem. In the context of radiology, an algorithm is usually a piece of computer code that takes a medical image as input and returns an answer to help the radiologist with his/her diagnosis.

  • To build an algorithm you (almost) always need a dataset, the training data, to get started. This dataset will be a batch of the type of data you want your algorithm to analyze. In radiology, this would be image data. Depending on the type of algorithm used, you may need additional information as well. This may be information on what you see in the image (e.g. a segmentation) or other patient information.

  • Depending on the context, several definitions for artificial intelligence can be used. Many of these definitions link human behavior to the (intended) behavior of a computer. In the case of radiology these definitions do not quite cover the scope of AI as there are many situations where AI exceeds human capabilities. In radiogenomics, for example, we link genetic information to what we see on medical images, enabling us to predict the presence or absence of genetic mutations in a tumor which can be used to determine further diagnosis and management.

  • Machine learning algorithms are a subset of artificial intelligence methods, characterized by the fact that you do not have to tell the computer how to solve the problem in advance. Instead, the computer learns to solve tasks by recognizing patterns in the data.

  • Deep learning is a subset of machine learning with the main differentiating factor being that deep learning uses “deep neural networks”, whereas machine learning comprises a much broader set of techniques. Deep neural networks are similar to the simple network described previously. However, deep networks have hidden layers between the input and the output layer to refine the calculations and hence the predictions. Simple neural networks require pre-processing to derive the image features which will be the input data for the network, whereas deep neural networks can use the image directly as input.

  • Radiologists are extremely busy healthcare professionals. They cannot afford to make any mistakes. They need to interact with a wide range of referring physicians; neurologists, urologists, orthopedic practitioners, the list goes on. They need to be sharp, always. What AI can bring these stretched radiologists and make them even better at what they do is:

    Provide a more differentiated diagnosis

    Pick up repetitive routine tasks, Offer a second opinion

    Eliminate inter- and intra-observer variability, Quality increase for better patient outcomes, Efficiency improvement to benefit the patient

  • No, it will not take over radiologist jobs. It most certainly will take over some radiologist tasks. It will support radiologists by performing automatic measurements which are currently very time consuming. It will pick up routine tasks which are experienced as cumbersome by a lot of radiologists. Yet, radiologists have a much more differentiated job than these type of tasks alone. In summary, AI is here to assist radiologist to make their lives easier.

Contact

Our Address

Woodstock Exchange, Albert Rd, Cape Town, 7915, South Africa