Artificial intelligence in pharma & drug development

Early detection and identification, high precision, reproducible data, automated analysis and tracking of treatment effects.

Find more about artificial intelligence in drug development.

AI in diagnostic imaging

By providing more precise and sensitive measures of drug efficacy and safety, AI in diagnostic imaging has the potential to substantially accelerate the drug discovery and development process. At GLI we combine state-of-the-art AI and experience in working with clinical experts. We’re sure that machine learning analysis of image data enables you to gain smarter insights and make more informed decisions.

Video & image analysis and processing

In our pharma-related projects, we always take advantage of novel image and video processing techniques to enhance, interpret, synthesize, translate, segment and analyze medical data. Our scientific computing team works every day on the best possible performance of our solutions combining convolutional neural networks, generative adversarial networks (GANs), Machine Learning and Deep Learning algorithms (Python & PyTorch).

Oncology clinical trials

Artificial intelligence has shown great potential in the field of oncological image analysis. We have successfully implemented solutions that identify, measure and track lesions across time points as well as provide radiomics features analysis. Leverage the experience of our data engineers and ML experts, and maximize the value of the analysis of medical imaging data in your pharma research and early development.

Imaging biomarkers development

Quantitative measurements derived from medical images can provide information about the biological processes underlying disease or the effects of a drug, and also to assess safety (safety biomarkers). We are experienced in working with clinical study teams on clinical datasets and datasets samples to identify imaging features that are predictive of disease progression, response to treatment, or other clinically relevant outcomes.

Artificial intelligence in pharma & drug development

Early detection and identification, high precision, reproducible data, automated analysis and tracking of treatment effects.

Find more about artificial intelligence in drug development.

AI in diagnostic imaging

By providing more precise and sensitive measures of drug efficacy and safety, AI in diagnostic imaging has the potential to substantially accelerate the drug discovery and development process. At GLI we combine state-of-the-art AI and experience in working with clinical experts. We’re sure that machine learning analysis of image data enables you to gain smarter insights and make more informed decisions.

Video & image analysis and processing

In our pharma-related projects, we always take advantage of novel image and video processing techniques to enhance, interpret, synthesize, translate, segment and analyze medical data. Our scientific computing team works every day on the best possible performance of our solutions combining convolutional neural networks, generative adversarial networks (GANs), Machine Learning and Deep Learning algorithms (Python & PyTorch).

Oncology clinical trials

Artificial intelligence has shown great potential in the field of oncological image analysis. We have successfully implemented solutions that identify, measure and track lesions across time points as well as provide radiomics features analysis. Leverage the experience of our data engineers and ML experts, and maximize the value of the analysis of medical imaging data in your pharma research and early development.

Imaging biomarkers development

Quantitative measurements derived from medical images can provide information about the biological processes underlying disease or the effects of a drug, and also to assess safety (safety biomarkers). We are experienced in working with clinical study teams on clinical datasets and datasets samples to identify imaging features that are predictive of disease progression, response to treatment, or other clinically relevant outcomes.

FOR PHARMA RESEARCH

Medical imaging analysis for early development and clinical trials

AI and machine learning in clinical trials

Machine learning and AI-enabled image analysis in drug development and the early development phase show great potential for considerably improving and simplifying radiological evaluation.

The example presents our models which automatically perform brain tumor segmentation and segment 3 tumor subregions (edema, enhancing tumor and necrosis) as well as provide volume and RANO (Response assessment in neuro-oncology) measurements. It can be a useful tool for monitoring patients in clinical studies.

We have been working with clinical study teams for several years, now we’d like to assist you with cutting-edge techniques.

INNOVATION IN PHARMA

Machine learning in pharma & drug development 

Imaging biomarkers development

Developing reliable imaging biomarkers can be a complex and time-consuming process but it’s worth it. Imaging readouts can be used to evaluate various aspects of a drug safety profile as well as to support the outcomes of your studies. Extract more value from your clinical datasets by leveraging our scientific and domain expertise.

Interactive graphics – rotate, zoom in, zoom out.

Talk with us today

Interactive graphics – rotate, zoom in, zoom out.

AI diagnostic imaging

Artificial intelligence may boost your medical data statistical power by limiting interobserver variability and increasing reproducibility.

In the example, we present our models for liver and liver tumor segmentation (including primary tumors, secondary tumors and metastases) based on portal venous CT scans. For clarity, only two biggest lesions are marked in the visualizations.

Talk to us about how AI-enabled diagnostic imaging and radiomics can nurture the consistency and efficiency of your clinical datasets.

Video and image analysis and processing

Artificial intelligence in drug development can be applied not only to DICOM images derived from CT, MRI and other examinations but also to video – we’ve been working on different endoscopy videos, providing our pharmaceutical partners with more insight into their datasets and even dataset samples. We combine cutting-edge techniques like Ml, DL, radiomics, as well as convolutional and generative adversarial networks (GAN) to work on the most demanding challenges, including low-dose and virtual non-contrast imaging.

Learn more about our projects for Pharma

In the below paper, we addressed the problem of detection of cirrhosis from CT, which is user-dependent and lacks reproducibility, and proposed an AI-enabled, reproducible end-to-end approach. Additionally, we showed that selecting the most discriminative features leads to the Pareto-optimal models with enhanced feature-level interpretability as the number of features was dramatically reduced (280) from thousands to tens.

The high performance of our algorithms is well presented in the below paper. Quantitative, qualitative and statistical analysis revealed that our pipeline performs accurate segmentation of pre and post-operative MRIs in a fraction of the manual delineation time (up to 20 times faster than humans). Additionally, we showed that RANO measurements are not always sufficient to quantify tumor burden and developed a new method of calculating RANO that is more resistant to poor automated segmentations and jagged contouring.

We tackled these issues in the paper below. We addressed the problem of detection of cirrhosis from CT, which is user-dependent and lacks reproducibility and proposed an AI-enabled, reproducible end-to-end approach. Additionally, we showed that selecting the most discriminative features leads to the Pareto-optimal models with enhanced feature-level interpretability, as the number of features was dramatically reduced (280) from thousands to tens.

The high performance of our algorithms is well presented in the below paper. Quantitative, qualitative and statistical analysis revealed that our pipeline performs accurate segmentation of pre- and post-operative MRIs in a fraction of the manual delineation time (up to 20 times faster than humans). Additionally, we showed that RANO measurements are not always sufficient to quantify tumor burden and developed a new method of calculating RANO that is more resistant to poor automated segmentations and jagged contouring.

Let’s work on your challenges together!

Contact us:

Let’s work on your challenges together!

Contact us: