MétaCan
Menu
← Back to cohort
Record W4409917909 · doi:10.1101/2025.04.25.650551

Performance of a Modular Robotic Cluster Matches Skilled Human Operators for Complex Cell Therapy Manufacturing Tasks

2025· preprint· en· W4409917909 on OpenAlexaff
Sudeshna Sadhu, Brigitte Schmittlein, Angela Lares, Sunaina Nayak, Marco Uboldi, Yifan Yu, Jiawei Wang, Wei Zhang, Prajakta Bhanap, Yongchang Ji, Andrew Scheffler, John F. Wilson, Dan Welch, Nikolaos Gkitsas-Long, Aidan Retherford, Ramya Tunuguntla, Alice Melocchi, Federico Parietti, Steven A. Feldman, Jonathan H. Esensten

Bibliographic record

VenuebioRxiv (Cold Spring Harbor Laboratory) · 2025
Typepreprint
Languageen
FieldMedicine
TopicCAR-T cell therapy research
Canadian institutionsUniversité de Sherbrooke
Fundersnot available
KeywordsComputer scienceManufacturing engineeringArtificial intelligenceEngineering

Abstract

fetched live from OpenAlex

Abstract Although numerous cell and gene therapies have received regulatory approval, their adoption has been hampered by high cost and challenges in scaling out manufacturing. Many autologous cell therapies are individually manufactured for each patient using traditional manual methods in high-cost environments. Therefore, robotics and automation offer a potential solution to meet the growing demand for such therapies. To ensure identical biological outcomes, automation must replicate validated manual workflows, a requirement that poses significant engineering challenges, especially for aseptic manipulation and compatibility with manual-centric instruments and consumables. Here we describe the design and performance of a modular robotic cluster consisting of specialized modules containing widely adopted equipment. The robotic arm uses custom end-effectors to handle standard consumables and instruments such as syringes, vials, bags, cell counters, bioreactors, incubators, and closed centrifuges. We compared the performance of skilled human operators against the robotic cluster across multiple tasks: transferring cells between sterile bags, cell counting, drawing volume from a vial to a syringe, and resuspension and sampling from both a bag and G-Rex 100M-CS bioreactor. The robotic system also executed high-complexity operations with industry-standard instruments: cell selection using a CytoSinct 1000 and wash/buffer exchange with a CTS Rotea Counterflow centrifugation system. Experimental results for each unit operation show that the robotic cluster’s performance is equivalent to manual operations on the selected key metrics. These data demonstrate that the robotic system can efficiently and robustly perform specific unit operations, which can be combined in any order for end-to-end cell therapy manufacturing processes. One Sentence summary A modular robotic system can perform key unit operations in cell therapy manufacturing with accuracy comparable to human operators while ensuring throughput and flexibility.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.002
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: Bench or experimental
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.004
Threshold uncertainty score0.014

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.002
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.000
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0010.001
Research integrity0.0010.000
Insufficient payload (model declined to judge)0.0040.002

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.

Opus teacher head0.032
GPT teacher head0.282
Teacher spread0.250 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venuebioRxiv (Cold Spring Harbor Laboratory)→Same topicCAR-T cell therapy research→French-language works237,207→