arXiv subjects

James Thomas

Publications and source records attributed to James Thomas.

14 recordsLinked to original sources

arXiv2026

Quality Action Assurance: Multimodal Verification of Examiner Claims in VR OSCEs

Objective Structured Clinical Examinations (OSCEs) are the gold standard for assessing clinical competence, yet scoring remains vulnerable to examiner subjectivity, fatigue, and cognitive bias. Standard examiner validation via inter-rater statistics lacks explanatory power regarding the source of errors, as it neither analyzes examiner reasoning nor verifies examiner claims against actual events. Thus, we introduce Quality Action Assurance (QAA), a multimodal framework that verifies examiner claims in Virtual Reality (VR) pediatric OSCEs by comparing actions claimed by examiners against a reference record of events constructed from video, VR logs, and actor annotations. QAA combines a constrained temporal action alignment model, which performs action localization and actor source attribution, with a large language model that extracts examiner claims and checks them against the record. Across a 5-fold cross-validation, QAA achieves 99.2\% $\pm$ 0.7\% Actor F1 and 93.4\% $\pm$ 1.9\% W@16 for temporal alignment. Overall, QAA detects examiner errors with 69.9\% precision and 76.7\% recall; in retrospective evaluation, correcting the detected errors raises the share of factually correct transcripts from 39.2\% to 79.2\%, supporting fairer OSCE quality assessment.

cs.AI↗

arXiv2025

Estimation of Resistance Training RPE using Inertial Sensors and Electromyography

Accurate estimation of rating of perceived exertion (RPE) can enhance resistance training through personalized feedback and injury prevention. This study investigates the application of machine learning models to estimate RPE during single-arm dumbbell bicep curls, using data from wearable inertial and electromyography (EMG) sensors. A custom dataset of 69 sets and over 1000 repetitions was collected, with statistical features extracted for model training. Among the models evaluated, a random forest classifier achieved the highest performance, with 41.4% exact accuracy and 85.9% $\pm1$ RPE accuracy. While the inclusion of EMG data slightly improved model accuracy over inertial sensors alone, its utility may have been limited by factors such as data quality and placement sensitivity. Feature analysis highlighted eccentric repetition time as the strongest RPE predictor. The results demonstrate the feasibility of wearable-sensor-based RPE estimation and identify key challenges for improving model generalizability.

cs.LG↗

arXiv2024

Exploring the use of a Large Language Model for data extraction in systematic reviews: a rapid feasibility study

This paper describes a rapid feasibility study of using GPT-4, a large language model (LLM), to (semi)automate data extraction in systematic reviews. Despite the recent surge of interest in LLMs there is still a lack of understanding of how to design LLM-based automation tools and how to robustly evaluate their performance. During the 2023 Evidence Synthesis Hackathon we conducted two feasibility studies. Firstly, to automatically extract study characteristics from human clinical, animal, and social science domain studies. We used two studies from each category for prompt-development; and ten for evaluation. Secondly, we used the LLM to predict Participants, Interventions, Controls and Outcomes (PICOs) labelled within 100 abstracts in the EBM-NLP dataset. Overall, results indicated an accuracy of around 80%, with some variability between domains (82% for human clinical, 80% for animal, and 72% for studies of human social sciences). Causal inference methods and study design were the data extraction items with the most errors. In the PICO study, participants and intervention/control showed high accuracy (>80%), outcomes were more challenging. Evaluation was done manually; scoring methods such as BLEU and ROUGE showed limited value. We observed variability in the LLMs predictions and changes in response quality. This paper presents a template for future evaluations of LLMs in the context of data extraction for systematic review automation. Our results show that there might be value in using LLMs, for example as second or third reviewers. However, caution is advised when integrating models such as GPT-4 into tools. Further research on stability and reliability in practical settings is warranted for each type of data that is processed by the LLM.

cs.CL↗

arXiv2021

Enabling Reusable Physical Design Flows with Modular Flow Generators

Achieving high code reuse in physical design flows is challenging but increasingly necessary to build complex systems. Unfortunately, existing approaches based on parameterized Tcl generators support very limited reuse and struggle to preserve reusable code as designers customize flows for specific designs and technologies. We present a vision and framework based on modular flow generators that encapsulates coarse-grain and fine-grain reusable code in modular nodes and assembles them into complete flows. The key feature is a flow consistency and instrumentation layer embedded in Python, which supports mechanisms for rapid and early feedback on inconsistent composition. The approach gradually types the Tcl language and allows both automatic and user-annotated static assertion checks. We evaluate the design flows of successive generations of silicon prototypes designed in TSMC16, TSMC28, TSMC40, SKY130, and IBM180 technologies, showing how our approach can enable significant code reuse in future flows.

cs.AR↗

arXiv2020

Two low-power optical data transmission ASICs for the ATLAS Liquid Argon Calorimeter readout upgrade

A serializer ASIC and a VCSEL driver ASIC are needed for the front-end optical data transmission in the ATLAS liquid argon calorimeter readout phase-I upgrade. The baseline ASICs are the serializer LOCx2 and the VCSEL driver LOCld, designed in a 0.25-{\mu}m Silicon-on-Sapphire (SoS) CMOS technology and consumed 843 mW and 320 mW, respectively. Based on a 130-nm CMOS technology, we design two pin-to-pin-compatible backup ASICs, LOCx2-130 and LOCld-130. Their power consumptions are much lower then of their counterparts, whereas other performance, such as the latency, data rate, and radiation tolerance, meet the phase-I upgrade requirements. We present the design of LOCx2-130 and LOCld-130. The test results of LOCx2-130 are also presented.

physics.ins-det↗

arXiv2020

The Latency Validation of the Optical Link for the ATLAS Liquid Argon Calorimeter Phase-I Trigger Upgrade

Two optical data link data transmission Application Specific Integrated Circuits (ASICs), the baseline and its backup, have been designed for the ATLAS Liquid Argon (LAr) Calorimeter Phase-I trigger upgrade. The latency of each ASIC and that of its corresponding receiver implemented in a back-end Field-Programmable Gate Array (FPGA) are critical specifications. In this paper, we present the latency measurements and simulation of two ASICs. The measurement results indicate that both ASICs achieve their design goals and meet the latency specifications. The consistency between the simulation and measurements validates the ASIC latency characterization.

physics.ins-det↗

arXiv2020

LOCx2-130, a low-power, low-latency, 2 x 4.8-Gbps serializer ASIC for detector front-end readout

In this paper, we present the design and test results of LOCx2-130, a low-power, low-latency, dual-channel transmitter ASIC for detector front-end readout. LOCx2-130 has two channels of encoders and serializers, and each channel operates at 4.8 Gbps. LOCx2-130 can interface with three types of ADCs, an ASIC ADC and two COTS ADCs. LOCx2-130 is fabricated in a commercial 130-nm CMOS technology and is packaged in a 100-pin QFN package. LOCx2-130 consumes 440 mW and achieves a latency of less than 40.7 ns.

physics.ins-det↗

arXiv2020

Compositions with 3 Pairwise Coprime Parts

How many ways can we write $n$ as a sum of $3$ positive integers, no pair of which share a common factor? We express this quantity in terms of the number of solutions to a certain class of linear Diophantine equations. This allows us to show that there are $$ \prod_{p \mid n} \left( 1- \frac{1}{p^2} \right) \prod_{q \nmid n} \left( 1- \frac{3}{q^2} \right) \frac{n^2}{2} + O(n^{3/2+o(1)}) $$ such compositions, where the products are over primes that respectively do and don't divide $n$. This strengthens the previous result of Bubbolini, Luca, and Spiga (arXiv:1202.1670)

math.NT↗

arXiv2019

Constructing Artificial Data for Fine-tuning for Low-Resource Biomedical Text Tagging with Applications in PICO Annotation

Biomedical text tagging systems are plagued by the dearth of labeled training data. There have been recent attempts at using pre-trained encoders to deal with this issue. Pre-trained encoder provides representation of the input text which is then fed to task-specific layers for classification. The entire network is fine-tuned on the labeled data from the target task. Unfortunately, a low-resource biomedical task often has too few labeled instances for satisfactory fine-tuning. Also, if the label space is large, it contains few or no labeled instances for majority of the labels. Most biomedical tagging systems treat labels as indexes, ignoring the fact that these labels are often concepts expressed in natural language e.g. `Appearance of lesion on brain imaging'. To address these issues, we propose constructing extra labeled instances using label-text (i.e. label's name) as input for the corresponding label-index (i.e. label's index). In fact, we propose a number of strategies for manufacturing multiple artificial labeled instances from a single label. The network is then fine-tuned on a combination of real and these newly constructed artificial labeled instances. We evaluate the proposed approach on an important low-resource biomedical task called \textit{PICO annotation}, which requires tagging raw text describing clinical trials with labels corresponding to different aspects of the trial i.e. PICO (Population, Intervention/Control, Outcome) characteristics of the trial. Our empirical results show that the proposed method achieves a new state-of-the-art performance for PICO annotation with very significant improvements over competitive baselines.

cs.CL↗

arXiv2018

Structured Multi-Label Biomedical Text Tagging via Attentive Neural Tree Decoding

We propose a model for tagging unstructured texts with an arbitrary number of terms drawn from a tree-structured vocabulary (i.e., an ontology). We treat this as a special case of sequence-to-sequence learning in which the decoder begins at the root node of an ontological tree and recursively elects to expand child nodes as a function of the input text, the current node, and the latent decoder state. In our experiments the proposed method outperforms state-of-the-art approaches on the important task of automatically assigning MeSH terms to biomedical abstracts.

cs.IR↗

arXiv2018

Permutations With Equal Orders

Let $P(n)$ be the probability that two independent, uniformly random permutations of $[n]$ have the same order, and let $K(n)$ be the probability that they are in the same conjugacy class. Answering a question of Thibault Godin, we prove that $ P(n)=n^{-2+o(1)}$ and that $\lim\sup \frac{ P(n) }{ K(n) }=\infty.$

math.CO↗

arXiv2018

LOCld65, A Dual-Channel VCSEL Driver ASIC For Detector Front-End Readout

We present the design and the test results of a dual-channel Vertical-Cavity Surface-Emitting Laser (VCSEL) driver ASIC LOCld65 for detector front-end readout. LOCld65 is designed in a commercial 65-nm CMOS technology with a power supply of 1.2 V. LOCld65 contains two separate channels with the same structure and the two channels share an I2C slave. Each channel consists of an input amplifier, four stages of limiting amplifiers (LAs), a high-current output driver, and a bias-current generator. In order to extend the bandwidth, the input amplifier uses an inductive peaking technique and the LAs use a shared inductive peaking technique. The input amplifier and the output driver each utilize a Continuous-Time Linear Equalizer (CTLE). The LAs employ active feedback. The modulation current, the bias current, the peaking strength of the CTLEs, and the feedback strength of LAs are programmable through an I2C interface. In order to protect from the radiation damage, the I2C slave is implemented with triple modular redundancy. Each channel of LOCld65 is tested to operate up to 14 Gbps with typical power dissipations (the VCSEL included) of 68.3 mW/channel and 62.1 mW/channel at the VCSEL voltages of 3.3 V and 2.5 V, respectively. LOCld65 survives 4.9 kGy(SiO2). LOCld65 is an excellent match for the serializer-deserializer ASIC lpGBT in single- or dual-channel optical transmitters in HL-LHC upgrade applications.

physics.ins-det↗

arXiv2018

Improving Active Learning in Systematic Reviews

Systematic reviews are essential to summarizing the results of different clinical and social science studies. The first step in a systematic review task is to identify all the studies relevant to the review. The task of identifying relevant studies for a given systematic review is usually performed manually, and as a result, involves substantial amounts of expensive human resource. Lately, there have been some attempts to reduce this manual effort using active learning. In this work, we build upon some such existing techniques, and validate by experimenting on a larger and comprehensive dataset than has been attempted until now. Our experiments provide insights on the use of different feature extraction models for different disciplines. More importantly, we identify that a naive active learning based screening process is biased in favour of selecting similar documents. We aimed to improve the performance of the screening process using a novel active learning algorithm with success. Additionally, we propose a mechanism to choose the best feature extraction method for a given review.

cs.IR↗

arXiv2017

Weld: Rethinking the Interface Between Data-Intensive Applications

Data analytics applications combine multiple functions from different libraries and frameworks. Even when each function is optimized in isolation, the performance of the combined application can be an order of magnitude below hardware limits due to extensive data movement across these functions. To address this problem, we propose Weld, a new interface between data-intensive libraries that can optimize across disjoint libraries and functions. Weld exposes a lazily-evaluated API where diverse functions can submit their computations in a simple but general intermediate representation that captures their data-parallel structure. It then optimizes data movement across these functions and emits efficient code for diverse hardware. Weld can be integrated into existing frameworks such as Spark, TensorFlow, Pandas and NumPy without changing their user-facing APIs. We demonstrate that Weld can speed up applications using these frameworks by up to 29x.

cs.DC↗