LoadingLoading
2020-07-13 15:27

Quality data is at the heart of AI medical diagnosis

Quality data is at the heart of AI medical diagnosis

The use of artificial intelligence models in healthcare continues to rise in demand as it enables medical professionals to make faster, early, and more accurate decisions that could save lives; such as in the case of early cancer detection. As accuracy is critical to prevent wrong diagnoses, Kili Technology can play a vital role in creating trustworthy and well labeled data, safely, smoothly and quickly. 

Book a demo

Success stories in healthcare

  • Scaling up AI for breast cancer screening
    Kili has enabled an American scaleup company to improve a model of assisted mammography diagnosis. The algorithm has been further enhanced to accurately interpret mammograms, enabling reliable detection of breast cancer at the level of the best experts.

  • Enabling innovative bladder cancer detection
    Kili is used by a French startup which is developing a new diagnostic tool that is non-invasive and more effective than urinary cytology for the detection of bladder cancer. Kili makes it possible to accelerate the classification of microscope images and to perform semantic segmentation of cancer cells.

  • Improving productivity with automatic speech recognition
    Kili helped a speech-to-text solution developer to train a model on medical (semantic) data to speed up transcription of radiologist reports. The project was conducted in three parts: data sourcing, data annotation and data augmentation.

Ensuring accuracy and productivity of medical AI models 

Kili technology makes annotating medical images and reports quick and simple. Import DICOM 2D, 3D CT Scan or MRI data, classify, draw bounding boxes, create polygons or segments to identify suspicious spots on the skin, lesions, tumors and brain haemorrhages. Build your training datasets with highly customizable interfaces that allow you to combine tasks to improve productivity.

In today's world, peta bytes of medical data are digitized in various healthcare institutions, public hospitals, retirement homes, medical clinics, pathology laboratories, etc. Unfortunately, these 2D or 3D images, CT Scan or MRI data are often disordered and unstructured. Unlike standard transactional business data, patient data is not directly usable to build models with machine learning.

These medical records must be annotated for the AI model to create impact. But manual data annotation can be expensive, laborious, and time-consuming. In addition, healthcare institutions need to pay extra attention to the quality of the annotated data, because the wrong diagnosis of patient data could dangerously affect people's lives.

Kili Technology can help by offering your teams the tools to create unbiased, trustworthy AI in one very cost-effective platform. We provide you with one central solution to data labeling that works all data types (including text, PDF, image, video, and audio) and addresses all use cases, from AI-assisted radiology and pathology to the identification of rare or difficult to diagnose diseases.

Create medical training datasets from data annotation
Many researchers around the world are looking to use computer vision models to detect skin cancer, brain tumors and other visually diagnosable diseases. However, creating and training these models requires access to large amounts of annotated medical image data.

It is not a major problem to find certain datasets. You can search for "medical datasets" in your favorite search engine. However, in order for a model to be able to make accurate predictions, it must be trained on a large amount of high-quality data that is specialized in the problem you want to address.

Thus, if you have to deal with a real use case, you will have no choice but to collect data very specific to your use case from a clinic or hospital and label it. Labeling can be expensive and prone to error. That's where Kili Technology comes in.


What makes Kili Technology different?

Expertise

Kili Technology manages DICOM 2D, 3D MRI or CT Scan images, and offers specialized interfaces for all annotation tasks related to intelligent document processing in healthcare. This includes medical imaging and NLP, image classification for visual diagnosis, identification of lesions, tumors, cancer cells, entity extraction for medical documents, OCR for medical records, and more.

Cost

Data annotation can be expensive. By allowing the use of online learning, active learning, weakly supervised learning or data augmentation, Kili Technology allows you to drastically reduce the cost of annotation.

Quality

Kili Technology provides best-in-class quality assurance management, which allows healthcare institutions to collaborate with ML experts to create trustworthy high-quality medical imaging training datasets, with less risk and more speed.  

Freedom

At Kili Technology, you can annotate wherever you want, with whomever you want. Kili can source and qualify the manpower you need to annotate on premises or offshore. We help you set clear guidelines and instructions so annotators know exactly what to do.

Scale

Kili Technology has access to a unique network of medical professionals around the world who are able to accurately translate, transcribe, and annotate medical data, so we can quickly create large, custom-built medical imaging and NLP training datasets at scale.

Training data interfaces for healthcare

Diagnostics for medical imaging

Add structure to the image with Bounding Box Annotation, Semantic Annotation, Polygon Annotation, Point Annotation, Segment Annotation, Image Classification, and more. We support the DICOM image format for AI in radiology.

Entity extraction for medical documents

Add structure and semantic information to unstructured text at the document and word level. Take advantage of our weakly supervised learning service to use business rules such as regular expressions and dictionaries to annotate massively before human intervention.

OCR for medical records

Crop parts of the text while saving the text to construct training data. Correct even the most subtle input errors, since even the smallest errors cannot be tolerated when handling sensitive medical data.

Last but not least, because our platform is as powerful as it is customizable, you can create your own interfaces to streamline processes, simplify collaboration, and ensure quality.

Get started

Get Started

Ready to simplify labeling in your company?