Kristin E. Lauter

Private predictive analysis on encrypted medical data

Increasingly, confidential medical records are being stored in data centers hosted by hospitals or large companies. As sophisticated algorithms for predictive analysis on medical data continue to be developed, it is likely that, in the future, more and more computation will be done on private patient data. While encryption provides a tool for assuring the privacy of medical information, it limits the functionality for operating on such data. Conventional encryption methods used today provide only very restricted possibilities or none at all to operate on encrypted data without decrypting it first. Homomorphic encryption provides a tool for handling such computations on encrypted data, without decrypting the data, and without even needing the decryption key. In this paper, we discuss possible application scenarios for homomorphic encryption in order to ensure privacy of sensitive medical data. We describe how to privately conduct predictive analysis tasks on encrypted data using homomorphic encryption. As a proof of concept, we present a working implementation of a prediction service running in the cloud (hosted on Microsofts Windows Azure), which takes as input private encrypted health data, and returns the probability for suffering cardiovascular disease in encrypted form. Since the cloud service uses homomorphic encryption, it makes this prediction while handling only encrypted data, learning nothing about the submitted confidential medical data.

Private Computation on Encrypted Genomic Data

A number of databases around the world currently host a wealth of genomic data that is invaluable to researchers conducting a variety of genomic studies. However, patients who volunteer their genomic data run the risk of privacy invasion. In this work, we give a cryptographic solution to this problem: to maintain patient privacy, we propose encrypting all genomic data in the database. To allow meaningful computation on the encrypted data, we propose using a homomorphic encryption scheme.

Homomorphic Computation of Edit Distance

These days genomic sequence analysis provides a key way of understanding the biology of an organism. However, since these sequences contain much private information, it can be very dangerous to reveal any part of them. It is desirable to protect this sensitive information when performing sequence analysis in public. As a first step in this direction, we present a method to perform the edit distance algorithm on encrypted data to obtain an encrypted result. In our approach, the genomic data owner provides only the encrypted sequence, and the public commercial cloud can perform the sequence analysis without decryption. The result can be decrypted only by the data owner or designated representative holding the decryption key.

HEALER: homomorphic computation of ExAct Logistic rEgRession for secure rare disease variants analysis in GWAS

MOTIVATIONnGenome-wide association studies (GWAS) have been widely used in discovering the association between genotypes and phenotypes. Human genome data contain valuable but highly sensitive information. Unprotected disclosure of such information might put individuals privacy at risk. It is important to protect human genome data. Exact logistic regression is a bias-reduction method based on a penalized likelihood to discover rare variants that are associated with disease susceptibility. We propose the HEALER framework to facilitate secure rare variants analysis with a small sample size.nnnRESULTSnWe target at the algorithm design aiming at reducing the computational and storage costs to learn a homomorphic exact logistic regression model (i.e. evaluate P-values of coefficients), where the circuit depth is proportional to the logarithmic scale of data size. We evaluate the algorithm performance using rare Kawasaki Disease datasets.nnnAVAILABILITY AND IMPLEMENTATIONnDownload HEALER at http://research.ucsd-dbmi.org/HEALER/ CONTACT: shw070@ucsd.edunnnSUPPLEMENTARY INFORMATIONnSupplementary data are available at Bioinformatics online.

Private genome analysis through homomorphic encryption

BackgroundThe rapid development of genome sequencing technology allows researchers to access large genome datasets. However, outsourcing the data processing o the cloud poses high risks for personal privacy. The aim of this paper is to give a practical solution for this problem using homomorphic encryption. In our approach, all the computations can be performed in an untrusted cloud without requiring the decryption key or any interaction with the data owner, which preserves the privacy of genome data.MethodsWe present evaluation algorithms for secure computation of the minor allele frequencies and χ2 statistic in a genome-wide association studies setting. We also describe how to privately compute the Hamming distance and approximate Edit distance between encrypted DNA sequences. Finally, we compare performance details of using two practical homomorphic encryption schemes - the BGV scheme by Gentry, Halevi and Smart and the YASHE scheme by Bos, Lauter, Loftus and Naehrig.ResultsThe approach with the YASHE scheme analyzes data from 400 people within about 2 seconds and picks a variant associated with disease from 311 spots. For another task, using the BGV scheme, it took about 65 seconds to securely compute the approximate Edit distance for DNA sequences of size 5K and figure out the differences between them.ConclusionsThe performance numbers for BGV are better than YASHE when homomorphically evaluating deep circuits (like the Hamming distance algorithm or approximate Edit distance algorithm). On the other hand, it is more efficient to use the YASHE scheme for a low-degree computation, such as minor allele frequencies or χ2 test statistic in a case-control study.