1 of 20

Privacy-Preserving Retrieval-Augmented Generation for Healthcare

IEEE HEALTHCOM 2026 · WCIHS WORKSHOP · NEW YORK, 19 TO 21 OCTOBER 2026

Two-Server Secret Sharing with Information-Theoretic Guarantees

Muntaser Syed, Ameerah Alsulami, Mahasen Enam, Nidadavolu Venkat Durga Sai Siva Vara Prasad Raju, Marius Silaghi

Florida Institute of Technology, Melbourne, FL, USA · Independent Researcher, UAE · Birla Institute of Technology and Science, Pilani, India

2 of 20

Healthcare RAG and Patient Privacy

INTRODUCTION

Why RAG in medicine

  • Grounds LLM answers in authoritative medical sources, reducing hallucination
  • Used for drug interactions, trial matching, radiology reports, medical QA
  • Retrieval quality sets response quality

Where PHI can leak

  • The query: a clinician's description can reveal diagnosis or treatment history
  • The corpus: clinical notes are sensitive by construction
  • HIPAA and GDPR apply; de-identified text is still re-identifiable

Goal: retrieval exactly as useful as today, with servers that learn nothing.

Syed et al. · IEEE HealthCom 2026

2 / 20

3 of 20

Existing Privacy Approaches

INTRODUCTION

Homomorphic encryption (CKKS)

CORRECT BUT SLOW

  • Approximate arithmetic adds noise to every operation
  • Megabyte-scale ciphertexts
  • Hundreds of milliseconds per query at best

Differential privacy

FAST BUT UNUSABLE

  • Calibrated noise added to the embeddings
  • Noise lands directly on the similarity scores
  • Retrieval fails at any meaningful privacy budget

Trusted execution environments

HARDWARE YOU MUST TRUST

  • Intel SGX, ARM TrustZone
  • Security rests on a hardware trust assumption
  • Repeatedly broken by side-channel attacks

Wanted: exact arithmetic, low latency, and no computational hardness assumption.

Syed et al. · IEEE HealthCom 2026

3 / 20

4 of 20

Split every embedding into two random shares. Neither server learns anything. Together they compute the exact inner product.

OUR APPROACH

Additive secret sharing over the 64-bit integer ring, with matrix-vector Beaver triples prepared offline, one per query. The servers reproduce the fixed-point inner product exactly. Security proofs and the triple construction are in the companion paper (Syed, Alsulami and Silaghi, IEEE ICTAI 2026).

Syed et al. · IEEE HealthCom 2026

4 / 20

5 of 20

Threat Model and Privacy Modes

APPROACH

Assumptions

  • Honest-but-curious servers: they follow the protocol but may learn from what they see
  • Two servers in separate domains; no collusion with each other or with the dealer
  • A dealer inside the hospital trust boundary prepares correlated randomness offline
  • Client and dealer trusted; TLS on every channel

Query-private mode

Only the query is protected; the knowledge base is public (drug references, guidelines).

Mutual-privacy mode

Query and corpus are both secret-shared. For sensitive clinical records.

Each server's view is independent of the protected embeddings; it learns only N, d, the precision, and the number and timing of queries.

Syed et al. · IEEE HealthCom 2026

5 / 20

6 of 20

Hospital and Cloud Deployment

APPROACH

1 Query shares to both servers

2 Masked query F exchanged (single-use mask)

3 Score shares to the client

4 Top-k from the local EHR

5 Docs and query to local LLM

6 Response to the clinician

The LLM never talks to the cloud server.

Syed et al. · IEEE HealthCom 2026

6 / 20

7 of 20

Offline Setup Phase

APPROACH

1 · Encode to fixed point

Multiply each coordinate by 2¹⁶ and round to the nearest integer. The largest inner product is about 2⁴¹, well inside the signed 64-bit range.

2 · Share the documents

Mutual-privacy: the dealer splits every document into two uniformly random shares, one per server. Query-private: documents stay in plaintext.

3 · Corpus tier (once per version)

The dealer shares a uniform mask A; the servers open E = D̂ - A once per corpus version. A changed document gets a fresh row of A first.

4 · Query tier (once per query)

The dealer pools single-use pairs (B, C = A·B); each query consumes one. Reusing a pair would reveal the difference between two queries.

Corpus tier: 0.5 to 114 ms for N ≤ 10,000; a pool of 1,000 query pairs is 79 MB per server.

Syed et al. · IEEE HealthCom 2026

7 / 20

8 of 20

Online Query Phase

APPROACH

1 · Share the query

The client encodes the query and sends one fresh random share to each server. Query-private mode: no Beaver material, no server-to-server traffic.

2 · Exchange the query mask

Mutual-privacy: each server broadcasts its share of F = q̂ - B for this query's single-use pair. 2d ring elements per query, independent of N.

3 · Compute score shares

Each server combines the public E and F with its private shares of A, B and C to form a share of every inner product.

4 · Reconstruct and rank

The client adds the two shares. Scores are exact fixed-point integers; cosine similarity is the score divided by 2³². Select the top-k.

Exact fixed-point arithmetic; agreement with float32 retrieval is empirical: 99.7 to 100%.

Syed et al. · IEEE HealthCom 2026

8 / 20

9 of 20

End-to-End RAG Pipeline

APPROACH

1

Embed

Client encodes the query with MedCPT or MiniLM

2

Encode and share

Fixed-point integers split into two fresh additive shares

3

Private retrieval

Two servers compute score shares without seeing the query or the shared documents

4

Reconstruct and rank

Client adds the shares, selects top-k, fetches plaintext from the local store

5

Secure generation

Local open-weight LLM (Llama 3, Meditron) or a BAA-covered endpoint

Generation stays inside the client's trust boundary. Sending PHI to a public LLM API would negate every guarantee of the protocol.

Syed et al. · IEEE HealthCom 2026

9 / 20

10 of 20

A single share reveals zero bits about the secret, regardless of the adversary's computational power.

PRIVACY ANALYSIS

Theorem 1 (informal, per mode). With honest-but-curious servers that collude with neither each other nor the dealer, each server's view is independent of the protected embeddings; it learns only N, d, the encoding precision, and the number and timing of queries.

Syed et al. · IEEE HealthCom 2026

10 / 20

11 of 20

HIPAA Technical Safeguards

PRIVACY ANALYSIS

Safeguard

Section

How the protocol meets it

Access Control

§164.312(a)

Each server holds only its share; no single server can reconstruct PHI.

Audit Controls

§164.312(b)

Share-level operation logs contain no PHI content.

Integrity

§164.312(c)

The secure computation adds no error; only the public, bounded fixed-point encoding.

Transmission Security

§164.312(e)

TLS in transit; the shares add a second layer of defense.

Minimum Necessary

§164.502(b)

No PHI-bearing content reaches either server beyond the public parameters.

Safe Harbor leaves data to re-identify; a single server's shares leave nothing.

Syed et al. · IEEE HealthCom 2026

11 / 20

12 of 20

Enforcing Non-Collusion

PRIVACY ANALYSIS

$2.19M

civil penalty cap per calendar year for violations of the same requirement (2025 inflation adjustment)

$250K

criminal fine, with up to 10 years of imprisonment (42 U.S.C. §1320d-6)

Both servers operate under separate Business Associate Agreements. Collusion means two organizations jointly violating independent legal obligations: a legal assumption rather than a hardness assumption, and one that holds against quantum computers.

Syed et al. · IEEE HealthCom 2026

12 / 20

13 of 20

Experimental Setup

EVALUATION

Datasets

PubMedQA (1,000 yes/no/maybe), MedQA (1,000 USMLE-style), MedMCQA (2,000 questions, 21 subjects). Corpora of 1,000 to 2,000 documents.

Embedding models

MedCPT, 768-d, trained on PubMed search logs. all-MiniLM-L6-v2, 384-d, as a general-purpose baseline.

Baselines

Plaintext float32 cosine similarity. CKKS homomorphic encryption at 128-bit security. Differential privacy: Gaussian noise, δ = 10⁻⁶, ε from 0.5 to 10, classical calibration.

Hardware and software

One workstation: RTX 4090, 64 GB RAM, Python 3.12. NumPy with native uint64 arithmetic. Every timing is a median over at least 50 repetitions.

Syed et al. · IEEE HealthCom 2026

13 / 20

14 of 20

Retrieval Agreement with Plaintext

EVALUATION

100%

Recall@1 and Recall@5 agreement

ρ = 1.0000

mean Spearman correlation, MPC

3.3 × 10⁻⁵

max similarity error (rounding)

Method

R@1

R@5

MPC-QP (ours)

1.000

1.000

MPC-MP (ours)

1.000

1.000

DP ε = 1

0.001

0.000

DP ε = 10

0.006

0.000

PubMedQA, 1,000 queries, 1,000 documents, MiniLM; MedCPT identical for MPC. R@k agreement = exact equality of the top-k sets.

At ε = 10, DP displaces the gold document by 325 to 393 ranks on average.

MedMCQA and MedQA: 99.7 to 100% Recall@k for MPC; DP at or below 0.2%.

Syed et al. · IEEE HealthCom 2026

14 / 20

15 of 20

End-to-End RAG Accuracy

EVALUATION

Retrieval

Accuracy

Yes

No

Maybe

Plaintext / MPC

0.430

0.562

0.215

0.391

DP ε = 1

0.205

0.116

0.185

0.696

DP ε = 10

0.180

0.089

0.138

0.739

Majority class (always yes)

0.560

1.000

0.000

0.000

PubMedQA, 200 queries (112 yes, 65 no, 23 maybe), top-5, Phi-3.5-mini (3.8B, Q4_K_M). Yes, No, Maybe = per-class accuracy. MPC returns the same top-5 sets as plaintext on every PubMedQA query (R@5 = 1.000).

43.0%

plaintext and MPC accuracy

76 to 78%

of DP answers are maybe

DP falls to 20.5% and 18.0%: below chance (33%) by the 95% intervals; McNemar p below 10⁻⁵.

Syed et al. · IEEE HealthCom 2026

15 / 20

16 of 20

Latency and Throughput

EVALUATION

1.43 ms

per query, query-private, N = 1,000, d = 384

698 q/s

throughput at the same point; 283 q/s in mutual-privacy mode

3.53 ms

mutual-privacy, incl. per-query triple generation; 1.4 to 3.2× QP

15.6 ms

query-private at N = 10,000; latency is linear in N and in d

Sub-millisecond up to N = 500 (QP, d = 384). Both servers in one process, no network.

Syed et al. · IEEE HealthCom 2026

16 / 20

17 of 20

Comparison with Baselines

EVALUATION

Method

R@1

Latency

Storage

Security

MPC-QP (ours)

1.000

1.4 ms

3.0 KB

IT

MPC-MP (ours)

1.000

3.5 ms

6.0 KB

IT

CKKS (N = 500)

1.000

7.7 s

327 KB

Comp.

DP ε = 1

0.001

not timed

3.0 KB

(ε, δ)

DP ε = 10

0.006

not timed

3.0 KB

(ε, δ)

N = 1,000, d = 384, MiniLM; CKKS measured at N = 500 (7,655 ms), not run at N = 1,000. Storage per document per server. IT = information-theoretic; Comp. = computational.

4,569 to 6,750×

faster than CKKS, mutual-privacy mode against CKKS at N = 100 and 500

27×

less storage per document: 12.0 KB of shares across both servers vs 326 KB; 54× per server

Syed et al. · IEEE HealthCom 2026

17 / 20

18 of 20

Limitations

CLOSING

01

Non-collusion is a trust assumption. The servers must collude with neither each other nor the dealer; enforceable through BAAs, absent from single-server HE.

02

Fixed-point encoding can swap near-ties. Score error at most 3.0 × 10⁻⁴ at d = 384 (3.3 × 10⁻⁵ observed); top-5 sets differed from float32 for at most 0.3% of queries.

03

Linear scan over all N documents. Practical at clinical corpus sizes; full-catalog retrieval (N above 10⁶) needs private approximate nearest neighbor pre-filtering.

04

Honest-but-curious servers; experimental randomness. SPDZ-style MACs extend to malicious adversaries at extra cost; deployment needs a CSPRNG in place of PCG64.

05

Model privacy is out of scope. The embedding model runs on the client. No real patient data was used; every evaluation runs on public benchmarks.

Syed et al. · IEEE HealthCom 2026

18 / 20

19 of 20

Conclusion

CLOSING

Exact secure computation

Bit-for-bit the fixed-point scores; top-5 agreement with float32 retrieval on 99.7 to 100% of queries across three benchmarks.

Information-theoretic security

Each server's view is independent of the protected embeddings; maps to the HIPAA technical safeguards and Minimum Necessary.

Millisecond latency

1.4 ms per query at 1,000 documents; three to four orders of magnitude faster than CKKS, with 27 to 54× less storage.

Deployable on existing infrastructure

Two servers under separate BAAs, a dealer and an LLM inside the hospital, and a FHIR-compatible interface at the client.

Future work: private approximate nearest neighbor search for full-catalog retrieval; evaluation on real clinical data under IRB approval; authenticated secret sharing against malicious servers.

Syed et al. · IEEE HealthCom 2026

19 / 20

20 of 20

Questions?

THANK YOU

Muntaser Syed · msyed2011@my.fit.edu

Florida Institute of Technology, Melbourne, FL, USA

Companion paper: M. Syed, A. Alsulami, and M. Silaghi, "Exact Privacy-Preserving Similarity Search via Two-Server Additive Secret Sharing," IEEE ICTAI 2026.

Syed et al. · IEEE HealthCom 2026

20 / 20