Development of a robotic orthosis for finger flexion motion by myoelectric control: open source prototype
Bibliographic record
Abstract
Neuromuscular disorders such as brachial plexus injury, stroke, and cerebral palsy affect the sensorimotor capabilities of the upper limbs. Individuals with such disorders have their performance in the most fundamental activities of daily living (ADLs) affected, and many are unable to return to work. This study proposed an open-source prototype of a myoelectric robotic hand orthosis to support ADLs. A “user-tuned” methodology was used, where the requirements recommended in recent literature and state of the art (2017 to 2020) were used as a starting point to design the orthosis’ system, usability, and ergonomics, but the prototype calibration for these requirements is performed using metrics taken from the user. The orthosis has two modules: (i) transmitter-interpreter system (TIS) and (ii) receiver-actuator system (RAS). The TIS is composed of a MyoWare surface electromyograph and an ESP32 imbued with a linear discriminant analysis (LDA) classifier trained with 10 datasets of surface myoelectric signals (sMES) recorded from a healthy volunteer. SMES are collected in real-time, segmented into 80 ms disruptive windows, filtered, and classified into three possible classes: (i) pulp pinch, (ii) transverse volar grip, and (iii) resting hand. A majority vote is cast on a vector containing 3 recent classified classes, and the class with the highest occurrence among the 3 is sent to the RAS. The entire TIS process occurs in 300 ms. The RAS module includes an ESP32, 2 N20 direct current motors, nylon artificial tendons, and wrist and finger splints printed in 3D technology. The RAS module receives a class from the TIS module and uses a finite state machine to decide which of the three hand poses to perform based on the class received, the current state of the machine, and the class previously received. Splints were constructed from measurements taken from the volunteer’s hand and adjusted until fitting and range of motion were comfortable for the volunteer. The test protocol was based on the Toronto Rehabilitation Institute Hand Function Test (TRI-HFT) and used everyday objects: (i) an ATM card mockup, (ii) a pencil, and (iii) a 600 ml plastic bottle filled with water. The prototype achieved all the ergonomics requirements recommended in the literature, and it was the orthosis that most fulfilled said requirements when compared to stateof-the-art myoelectric robotic hand orthoses. Additionally, the system’s accuracy reached 90% in real-time tests. Finally, the cost of producing the orthosis is R$856.35 or 2.98% of the value of the cheapest commercial myoelectric hand orthosis on the market (PowerGrip), both prices quoted in November 2021.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.011 | 0.003 |
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".