Synthetic Data Generation By Artificial Intelligence to Accelerate Translational Research and Precision Medicine in Hematological Malignancies

D'Amico, Saverio; Sauta, Elisabetta; Bersanelli, Matteo; Dall’Olio, Daniele; Sala, Claudia; Dall’Olio, Lorenzo; Morandini, Pierandrea; Tommasini, Tobia; Bicchieri, Marilena; Zampini, Matteo; Savevski, Victor; Prada-Luengo, Iñigo; Krogh, Anders; Platzbecker, Uwe; Díez-Campelo, María; Fenaux, Pierre; Haferlach, Torsten; Gastone, Castellani; Porta, Matteo Giovanni Della

doi:10.1182/blood-2022-168646

Cited by 6 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some experiences in the same or other settings have been recently published. Though employing different methods to generate synthetic patients, they witness the increasing interest and potential of synthetic data [14,15].…”

Section: Discussionmentioning

confidence: 99%

Unlocking the potential of synthetic patients for accelerating clinical trials: Results of the first GIMEMA experience on acute myeloid leukemia patients

Piciocchi,

Cipriani,

Messina

et al. 2024

eJHaem

View full text Add to dashboard Cite

Artificial Intelligence has the potential to reshape the landscape of clinical trials through innovative applications, with a notable advancement being the emergence of synthetic patient generation. This process involves simulating cohorts of virtual patients that can either replace or supplement real individuals within trial settings. By leveraging synthetic patients, it becomes possible to eliminate the need for obtaining patient consent and creating control groups that mimic patients in active treatment arms. This method not only streamlines trial processes, reducing time and costs but also fortifies the protection of sensitive participant data. Furthermore, integrating synthetic patients amplifies trial efficiency by expanding the sample size. These straightforward and cost‐effective methods also enable the development of personalized subject‐specific models, enabling predictions of patient responses to interventions. Synthetic data holds great promise for generating real‐world evidence in clinical trials while upholding rigorous confidentiality standards throughout the process. Therefore, this study aims to demonstrate the applicability and performance of these methods in the context of onco‐hematological research, breaking through the theoretical and practical barriers associated with the implementation of artificial intelligence in medical trials.

show abstract

Section: Discussionmentioning

confidence: 99%

Unlocking the potential of synthetic patients for accelerating clinical trials: Results of the first GIMEMA experience on acute myeloid leukemia patients

Piciocchi,

Cipriani,

Messina

et al. 2024

eJHaem

View full text Add to dashboard Cite

show abstract

“…65 AI/ML models for SDG have shown to emulate real data characteristics in various therapeutic areas, including but not limited to hematology, oncology, infectious diseases, medical imaging, and endocrinology for lucrative applications, such as estimation of treatment effect and survival, use as a proxy for clinical trial datasets to perform secondary analyses, generate large datasets for development of image segmentation models, predict future patient outcomes (e.g., glycemic change), and data augmentation to develop disease diagnosis models. [66][67][68][69][70] Additionally, open source tools, such as Synthea, have been developed for generating EHRs data of patient disease progression and clinical workflow. 71 These examples and open-source software highlight the utility of synthetic data while maintaining privacy of patient data.…”

Section: Data Sharing Distributed Learning and Synthetic Data Generationmentioning

confidence: 99%

“…Use of synthetic data can not only help preserve patient privacy but also help generate hypotheses in the process of obtaining real datasets, augment real datasets (partial synthetic data), ease sharing data to verify analyses and improve reproducibility, and pre‐train models to be used for application in specific populations 65 . AI/ML models for SDG have shown to emulate real data characteristics in various therapeutic areas, including but not limited to hematology, oncology, infectious diseases, medical imaging, and endocrinology for lucrative applications, such as estimation of treatment effect and survival, use as a proxy for clinical trial datasets to perform secondary analyses, generate large datasets for development of image segmentation models, predict future patient outcomes (e.g., glycemic change), and data augmentation to develop disease diagnosis models 66–70 . Additionally, open source tools, such as Synthea, have been developed for generating EHRs data of patient disease progression and clinical workflow 71 .…”

Section: Challenges and Future Directionmentioning

confidence: 99%

Current Status and Future Directions: The Application of Artificial Intelligence/Machine Learning for Precision Medicine

Naik,

Goyal,

Foschini

et al. 2024

Clin Pharma and Therapeutics

View full text Add to dashboard Cite

Technological innovations, such as artificial intelligence (AI) and machine learning (ML), have the potential to expedite the goal of precision medicine, especially when combined with increased capacity for voluminous data from multiple sources and expanded therapeutic modalities; however, they also present several challenges. In this communication, we first discuss the goals of precision medicine, and contextualize the use of AI in precision medicine by showcasing innovative applications (e.g., prediction of tumor growth and overall survival, biomarker identification using biomedical images, and identification of patient population for clinical practice) which were presented during the February 2023 virtual public workshop entitled “Application of Artificial Intelligence and Machine Learning for Precision Medicine,” hosted by the US Food and Drug Administration (FDA) and University of Maryland Center of Excellence in Regulatory Science and Innovation (M‐CERSI). Next, we put forward challenges brought about by the multidisciplinary nature of AI, particularly highlighting the need for AI to be trustworthy. To address such challenges, we subsequently note practical approaches, viz., differential privacy, synthetic data generation, and federated learning. The proposed strategies – some of which are highlighted presentations from the workshop – are for the protection of personal information and intellectual property. In addition, methods such as the risk‐based management approach and the need for an agile regulatory ecosystem are discussed. Finally, we lay out a call for action that includes sharing of data and algorithms, development of regulatory guidance documents, and pooling of expertise from a broad‐spectrum of stakeholders to enhance the application of AI in precision medicine.

show abstract

“…The so-called "digital patients" are virtual copies of real patients which are created based on clinical and genomic data collected in real-world data sources. The generation of synthetic cohorts to be used as control arms holds promise to accelerate clinical research in those fields where the conduction of such studies is limited by ethical and/or feasibility issues, such as rare diseases and pediatric diseases in general (D'Amico et al, 2023).…”

Section: Artificial Intelligence For Drug Discovery and Repurposing C...mentioning

confidence: 99%

Artificial intelligence for optimizing benefits and minimizing risks of pharmacological therapies: challenges and opportunities

Crisafulli,

Ciccimarra,

Bellitto

et al. 2024

Front. Drug Saf. Regul.

View full text Add to dashboard Cite

In recent years, there has been an exponential increase in the generation and accessibility of electronic healthcare data, often referred to as “real-world data”. The landscape of data sources has significantly expanded to encompass traditional databases and newer sources such as the social media, wearables, and mobile devices. Advances in information technology, along with the growth in computational power and the evolution of analytical methods relying on bioinformatic tools and/or artificial intelligence techniques, have enhanced the potential for utilizing this data to generate real-world evidence and improve clinical practice. Indeed, these innovative analytical approaches enable the screening and analysis of large amounts of data to rapidly generate evidence. As such numerous practical uses of artificial intelligence in medicine have been successfully investigated for image processing, disease diagnosis and prediction, as well as the management of pharmacological treatments, thus highlighting the need to educate health professionals on these emerging approaches. This narrative review provides an overview of the foremost opportunities and challenges presented by artificial intelligence in pharmacology, and specifically concerning the drug post-marketing safety evaluation.

show abstract

Synthetic Data Generation By Artificial Intelligence to Accelerate Translational Research and Precision Medicine in Hematological Malignancies

Cited by 6 publications

References 0 publications

Unlocking the potential of synthetic patients for accelerating clinical trials: Results of the first GIMEMA experience on acute myeloid leukemia patients

Unlocking the potential of synthetic patients for accelerating clinical trials: Results of the first GIMEMA experience on acute myeloid leukemia patients

Current Status and Future Directions: The Application of Artificial Intelligence/Machine Learning for Precision Medicine

Artificial intelligence for optimizing benefits and minimizing risks of pharmacological therapies: challenges and opportunities

Contact Info

Product

Resources

About