February 2026: Shamir and Benesh: Improving the Parkinson's disease scale

A new study by Prof. Ron Shamir, Edmond J. Safra member, and his PhD student, Assaf Benesh (Computer Science & AI), on improving the standard progression scale for Parkinson's disease (MDS-UPDRS) was published in npj Parkinson's Disease.

The study is challenging long‑standing assumptions about the Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale (MDS‑UPDRS), the gold‑standard tool for assessing Parkinson’s disease. The MDS‑UPDRS contains 50 questions that aim to evaluate the severity of a patient’s disease across four parts: Non‑motor daily‑living experiences (sleep, mood, pain, autonomic issues), Patient‑reported motor daily‑living abilities, and Physician‑rated motor examination and motor complications. Each question is scored from 0 to 4, and the total score is calculated by simply summing all items.

The researchers hypothesized that the extent to which the scoring system reflects disease progression can be improved. They reasoned that the current MDS‑UPDRS treats all symptoms and score increments as equally important, which is simplistic. Certain questions may contribute little to tracking long‑term progression. Some items with the same numeric score represent very different clinical realities that have different contributions to disease progression.

The study, conducted in collaboration with Prof. Nir Giladi (z”l), Prof. Anat Mirelman, and Prof. Roy Alcalay of Tel Aviv Sourasky Medical Center (Ichilov), developed several mathematical optimization models for this challenge. Using these models, they reweighted questions and score steps and produced an index that is substantially more monotonic along disease time. Moreover, a much simpler progression index, using fewer items that are all self-reported by the patient, can be produced, almost without losing accuracy.

The team recommends reporting both the traditional MDS‑UPDRS score and the optimized index side‑by‑side. As Prof. Shamir explains in a news item published in The Movement Disorder Journal and News: "We do not wish to replace the MDS‑UPDRS… What we propose is to summarize it differently, in a way that reflects better the tendency of the disease to progress over time". The proposed optimized index could improve clarity, reduce noise in long‑term monitoring, and help clinicians track meaningful changes more effectively.

This study was supported by a joint grant program from the Center for Artificial Intelligence and Data Science at Tel Aviv University (TAD), The Edmond J. Safra Center for Bioinformatics, and Teva Pharmaceutical Industries Ltd.