Efficient Server-Aided Personalized Treatment Recommendation with Privacy Preservation
Bibliographic record
Abstract
With AI-derived knowledge graph (KG), medical centers can recommend appropriate treatment options to physicians as references based on their patients' personal healthcare information (PHI). However, the treatment recommendation services may also cause serious privacy concerns. In this paper, we propose an efficient and privacy-preserving personalized treatment recommendation scheme with the aid of a third-party server. Specifically, a medical center denotes the KG of each disease by a directed graph with conditional edges and vertices that describe the treatment options and costs in different states. To prevent privacy leakage while reducing computational and management cost, the graphs are encrypted and then delegated to an honest-but-curious server. With the assistance of the server, physicians can set proper illness states and cost requirements according to patients' PHI, and correspondingly generate personalized ciphertexts to retrieve appropriate treatment options. The key component of the proposed scheme is a novel designed secure and flexible path comparison protocol, by tailoring a symmetric homomorphic encryption algorithm and combining it with a secure hash function. The protocol can enable the server to compare the uploaded ciphertexts with encrypted graphs in a secure and efficient way. Comprehensive security analysis indicates that the proposed scheme can meet desirable privacy requirements, and extensive experimental results demonstrate its practicality.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".