SPuMoNI
Smart Pharmaceutical Manufacturing
EU-funded Research Project
Nominated Award:
Best Application of AI in an Academic Research Body
Website of Company (or Linkedin profile of Person):
http://www.spumoni.eu/
Funded by the European Commission from April 2019 until December 2022, the Smart Pharmaceutical Manufacturing Project (SPuMoNI) http://www.spumoni.eu harnesses the potential of state-of-the-art technologies for the pharmaceutical industry, such as blockchain for end-to-end verification of manufacturing data and modern AI and data analytics to smartly extract, transform, and control heterogeneous data sources within the manufacturing processes in order to better improve Big Data quality and process modelling for a smarter industry with emphasis on data quality assurance for data integrity compliance.
The SPuMoNI consortium comprises two pharma industrial partners, PQE Group and FAREVA’s Instituto De Angeli; and, three academic institutions, the Universidad Politécnica de Valencia (Spain), the University of Thessaly (Greece) and the National College of Ireland (Ireland).
Currently running in its latest stages and with a proof of concept already available upon request for demonstration, the SPuMoNI system delivers an ALCOA+ Assessment, the pharmaceutical industry gold standard for quality manufacturing, to ensure continuous data integrity of pharma manufacturing reports.
Furthermore, SPuMoNI has achieved the following significant results in the past three years:
• Collected anonymized datasets with discrete and specific attributes related to environmental conditions of different pharma systems, which are useful for the development of software that already structures data in this fashion.
• Issued data integrity guidelines to set the rules on how Artificial Intelligence (AI) should process data and identify patterns that may lead to compliance issues.
• Developed the base AI architecture. Further development of additional AI applications for other manufacturing processes related to the pharmaceutical industry, to assess data integrity compliance before validation/deployment.
• Enhanced multi-node private blockchain networks to ensure data provenance and compliance in a tamper-proof manner.
• Released the SPuMoNI guidelines as a template of integrated software/network infrastructure for pharma manufacturing.
• Deployed a prototype in an industrially relevant environment.
Reason for Nomination:
The EU Falsified Medicines Directive “introduces harmonised European measures to fight medicine falsifications and ensure that medicines are safe and that the trade in medicines is rigorously controlled”. Such obligatory safety features, legal framework, and record-keeping requirements have arguably imposed stricter controls for manufacturing of medicaments. While the pharmaceutical industry has consistently improved their manufacturing processes in compliance with good manufacturing practices , it is documented that deviation to good practices and even falsification of medicines continue.
Disclosure risk assessment techniques in pharma manufacturing typically depend on background knowledge, the behaviour of intruders and the specific value of the data. Often only heuristic arguments are used without numerical assessment . Thus, SPuMoNI leverages blockchain and AI technologies to better ascribe and ensure the manufacturing traceability of medicaments. Spumoni is particularly timely as blockchain has been recently proposed to become “a new Digital Service Infrastructure” for Europe . While blockchain is well-established in the cryptocurrency domain, the systematic application of smart contracts in the pharma industry remains an open problem. Moreover, traceability in manufacturing has traditionally been studied in the food industry, but rarely in the pharmaceutical manufacturing side consequently attracting some industry attention .
SPuMoNI instruments a compliant blockchain-based end-to end verification system for medicaments using applied AI-based quality assurance techniques, as required in EU member states by February 2019 (2025 for Belgium, Italy, and Greece) as part of the greater EU Falsified Medicines Directive initiated through the Delegated Acts on safety features (Regulation 2016/161).
To address this problem, SPuMoNI exploits and build on existing open software systems, frameworks and standards, such as OpenStack and Ethereum to develop and evaluate AI-based quality control mechanisms for pharmaceutical data specifically with respect to hinder data falsifiability. It also verifies that security, privacy, compliance and ownership concerns have been properly met for the distributed, possibly open big data, that is processed and generated within pharmaceutical manufacturing contexts and environments, while intelligently coordinating data gathering, and processing within a myriad of contexts and environments.
SPuMoNI has developed an adaptive approach to intelligent data analytics in pharma manufacturing employing computational agents embedding decentralised machine learning techniques on novel distributed processing platforms.
The applicability of the SPuMoNI AI technologies with respect to real-world manufacturing data and environments have been demonstrated in real pharma manufacturing environments. SPuMoNI is producing a technology roadmap for AI in pharma manufacturing and pursuing other innovation and exploitation activities to maximise market potential in the long term, including implementing an open data strategy to disseminate the project results.
Additional details of the project can be learned from the following peer-reviewed scientific articles, published as Open Access and therefore enclosed for completeness :
Fátima Leal, Adriana E. Chis, Simon Caton, Horacio González-Vélez, et al.: Smart Pharmaceutical Manufacturing: Ensuring End-to-End Traceability and Data Integrity in Medicine Production. Big Data Research. 24: 100172 (2021)
https://doi.org/10.1016/j.bdr.2020.100172
Marta Durá, Angel Sánchez-García, Carlos Sáez, Fátima Leal, Adriana E. Chis, Horacio González-Vélez, Juan M. García-Gómez: Towards a Computational Approach for the Assessment of Compliance of ALCOA+ Principles in Pharma Industry. Challenges of Trustable AI and Added-Value on Health – Proceedings of MIE 2022, Nice, France, May 27-30, 2022. 2022: 755-759 https://doi.org/10.3233/SHTI220578
Additional Information:
This nomination is put forward by Dr Horacio Gonzalez-Velez. As the founding head of NCI’s Cloud Competency Centre and associate professor of computing since 2012, Dr González—Vélez carries out research and innovation endeavours related to parallel and distributed computing with emphasis on programmability and parallel patterns. He held different positions in product marketing and HPC systems engineering at innovation-driven multinationals Silicon Graphics and Sun Microsystems from 1995 to 2003, where he had different opportunities to work with a number of dot-com SMEs in The Americas. Subsequently, he earned a PhD in Informatics from the University of Edinburgh in 2008 where he remained as a postdoctoral research fellow.
The National College of Ireland (NCI), https://www.ncirl.ie, is an Irish not-for-profit, state-aided third-level education institution founded in 1951. As of the end of academic year 2020-21, NCI has some 6,000 students in full and part-time undergraduate and postgraduate programmes including doctoral degrees, with over 2,000 registered in computing degrees. In 2016, NCI became the first institution outside of the Institute of Technology sector to offer higher education apprenticeships. NCI will participate in this project via The Cloud Competency Centre (CCC). From its inception within the School of Computing in 2012, CCC has brought forth tangible high-impact synergies across the world. With an emerging researcher base, CCC remains committed to bridging high-performance computing, distributed ledger technologies, data analytics, and parallel processing with ongoing business and economic developments to produce quality research outputs and collaborations with academic institutions, funding agencies, and companies.
CCC also offers three 90-credit MSc programmes in Cloud Computing, in Data Analytics, and in FinTech. In line with NCI’s overarching mission of widening access to education and the official agenda to build knowledge economy skills, CCC places a strong emphasis on leading aspects of teaching, applied research, and commercialisation in these three programmes including company-based and blended-learning activities. CCC has a dedicated Cloud resource portal available at https://cloud.ncirl.ie which includes access to its private cloud. Configured using the full OpenStack cloud software stack, this 2000-core private cloud has processing capabilities to concurrently support 1000 active instances, over 10 Terabytes of block/blob storage, and fast network links via HEAnet, Ireland’s National Education and Research Network.
Chaired by Professor Emeritus and ACM A.M. Turing Award 2020 winner Jeffrey D. Ullman from Stanford University, the CCC academic advisory board includes Professors Michael Franklin from the University of Chicago, Professor Armando Fox from the University of California-Berkeley, and ACM A.M. Turing Award 1986 winner Professor John Hopcroft from Cornell University. CCC is also collaborating closely with industry representatives from global ICT companies and indigenous Irish SMEs.
Nevertheless, the need for innovative postgraduate programmes such as the ones proposed in this project is clear across the sector. The FIT ICT Skills Audit for 2018 Report finds demand is strong across a wide range of tech disciplines with 12,000 current vacancies in the sector as a whole, offering quality job opportunities at entry, competent as well as expert levels, and for young and old alike.