Gabriele Spini1, Emiliano Mancini2,3,4, Thomas Attema5,6,7, Mark Abspoel6,8, Jan de Gier5, Serge Fehr6,7, Thijs Veugen5,6, Maran van Heesch5, Daniël Worm5, Andrea De Luca9, Ronald Cramer6,7, Peter M A Sloot2,10,11. 1. Applied Cryptography and Quantum Algorithms, TNO, 96800, 2509 JE, Postbus, The Hague, The Netherlands. gabriele.spini@tno.nl. 2. Institute for Advanced Study, University of Amsterdam, Oude Turfmarkt 147, 1012 GC, Amsterdam, The Netherlands. 3. Department of Global Health, Amsterdam UMC, Location AMC, 1105 AZ, Amsterdam, The Netherlands. 4. Data Science Institute, Hasselt University, Diepenbeek, Belgium. 5. Applied Cryptography and Quantum Algorithms, TNO, 96800, 2509 JE, Postbus, The Hague, The Netherlands. 6. Cryptology Group, CWI, P.O. Box 94079, 1090 GB, Amsterdam, The Netherlands. 7. Mathematical Institute, Leiden University, P.O. Box 9512, 2300 RA, Leiden, The Netherlands. 8. Philips Research, High Tech Campus 34, 5656 AE, Eindhoven, The Netherlands. 9. Department of Medical Biotechnologies, University of Siena and Siena University Hospital, Viale Mario Bracci 16, 53100, Siena, Italy. 10. Complexity Institute, Nanyang Technological University, Academic Building North, Level 1 Section B Unit No. 7 (ABN-01B-07), 61 Nanyang Drive, 637335, Singapore, Singapore. 11. Advanced Computing, ITMO University, Lomonosova street 9, 191002, Saint Petersburg, Russia.
Abstract
BACKGROUND: HIV treatment prescription is a complex process. Clinical decision support systems (CDSS) are a category of health information technologies that can assist clinicians to choose optimal treatments based on clinical trials and expert knowledge. The usability of some CDSSs for HIV treatment would be significantly improved by using the knowledge obtained by treating other patients. This knowledge, however, is mainly contained in patient records, whose usage is restricted due to privacy and confidentiality constraints. METHODS: A treatment effectiveness measure, containing valuable information for HIV treatment prescription, was defined and a method to extract this measure from patient records was developed. This method uses an advanced cryptographic technology, known as secure Multiparty Computation (henceforth referred to as MPC), to preserve the privacy of the patient records and the confidentiality of the clinicians' decisions. FINDINGS: Our solution enables to compute an effectiveness measure of an HIV treatment, the average time-to-treatment-failure, while preserving privacy. Experimental results show that our solution, although at proof-of-concept stage, has good efficiency and provides a result to a query within 24 min for a dataset of realistic size. INTERPRETATION: This paper presents a novel and efficient approach HIV clinical decision support systems, that harnesses the potential and insights acquired from treatment data, while preserving the privacy of patient records and the confidentiality of clinician decisions.
BACKGROUND: HIV treatment prescription is a complex process. Clinical decision support systems (CDSS) are a category of health information technologies that can assist clinicians to choose optimal treatments based on clinical trials and expert knowledge. The usability of some CDSSs for HIV treatment would be significantly improved by using the knowledge obtained by treating other patients. This knowledge, however, is mainly contained in patient records, whose usage is restricted due to privacy and confidentiality constraints. METHODS: A treatment effectiveness measure, containing valuable information for HIV treatment prescription, was defined and a method to extract this measure from patient records was developed. This method uses an advanced cryptographic technology, known as secure Multiparty Computation (henceforth referred to as MPC), to preserve the privacy of the patient records and the confidentiality of the clinicians' decisions. FINDINGS: Our solution enables to compute an effectiveness measure of an HIV treatment, the average time-to-treatment-failure, while preserving privacy. Experimental results show that our solution, although at proof-of-concept stage, has good efficiency and provides a result to a query within 24 min for a dataset of realistic size. INTERPRETATION: This paper presents a novel and efficient approach HIV clinical decision support systems, that harnesses the potential and insights acquired from treatment data, while preserving the privacy of patient records and the confidentiality of clinician decisions.
Authors: Jaideep Ravela; Bradley J Betts; Francoise Brun-Vézinet; Anne-Mieke Vandamme; Diane Descamps; Kristel van Laethem; Kate Smith; Jonathan M Schapiro; Dean L Winslow; Caroline Reid; Robert W Shafer Journal: J Acquir Immune Defic Syndr Date: 2003-05-01 Impact factor: 3.731
Authors: Peter M A Sloot; Alexander V Boukhanovsky; Wilco Keulen; Alfredo Tirado-Ramos; Charles A Boucher Journal: J Clin Monit Comput Date: 2005-10 Impact factor: 2.502
Authors: Peter Sloot; Peter Coveney; Marian T Bubak; Anne-Mieke Vandamme; Breanndán O Nualláin; David van de Vijver; Charles Boucher Journal: Stud Health Technol Inform Date: 2008
Authors: Charles B Hicks; Pedro Cahn; David A Cooper; Sharon L Walmsley; Christine Katlama; Bonaventura Clotet; Adriano Lazzarin; Margaret A Johnson; Dietmar Neubacher; Douglas Mayers; Hernan Valdez Journal: Lancet Date: 2006-08-05 Impact factor: 79.321
Authors: Michael A Horberg; Judith A Aberg; Laura W Cheever; Philip Renner; Erin O'Brien Kaleba; Steven M Asch Journal: Clin Infect Dis Date: 2010-09-15 Impact factor: 9.079
Authors: P M A Sloot; Peter V Coveney; G Ertaylan; V Müller; C A Boucher; M Bubak Journal: Philos Trans A Math Phys Eng Sci Date: 2009-07-13 Impact factor: 4.226
Authors: John R Beard; Alana Officer; Islene Araujo de Carvalho; Ritu Sadana; Anne Margriet Pot; Jean-Pierre Michel; Peter Lloyd-Sherlock; JoAnne E Epping-Jordan; G M E E Geeske Peeters; Wahyu Retno Mahanani; Jotheeswaran Amuthavalli Thiyagarajan; Somnath Chatterji Journal: Lancet Date: 2015-10-29 Impact factor: 79.321