Linhai Ma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Linhai Ma has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
Afshan Khan
2 shared publicationsGanesh Puthiaraju
2 shared publications- Samah Jamal Fodeh
Yale University, New Haven, USA
2 shared publications Sarah Schellhorn
2 shared publications- Srivani Talakokkul
Department of Emergency Medicine, Yale School of Medicine, 464 Congress Ave, New Haven, 06519, CT, USA
2 shared publications Aimee Roundtree
Division of Research Texas State University 601 University Dr. 78666 TX USA
1 shared publication
Latest Publications
- Open Access
- 25 September 2026
- Published by Elsevier in SSRN Electronic Journal
STaR-DRO: Stateful Tsallis Reweighting for Group-Robust Structured Prediction
Structured prediction with large language models requires more than accurate labels: outputs must satisfy ontology constraints, preserve valid structure, and ensure evidence-grounded decisions under label imbalance and heterogeneous group difficulty. To resolve these challenges, we present a unified framework for robust ontology-constrained generation. First, we introduce a modular prompt-engineering architecture combining XML-style structure, expert disambiguation rules, chain-of-thought reasoning, metadata-aware decision logic, schema contracts, and a self-validation gate. This targets recurrent in-context structured generation failures, including format drift, label ambiguity, evidence hallucination, and metadata-conditioned confusion. Second, we propose STaR-DRO, which combines Tsallis mirror ascent with sparse entmax-style primal mapback, EMA-smoothed group-loss tracking, rescaled ascent signals, and bounded excess-only multipliers. This upweights only persistently hard groups without suppressing easier ones, addressing conventional DRO’s dense Shannon-entropy exponentiated-gradient updates, high-variance stochastic reweighting, positive adversarial mass on groups that are not persistently hard, and performance costs from simplex competition. We evaluate the framework on EPPC Miner, a carefully chosen, clinically grounded representation of high-stakes structured prediction requiring hierarchical label prediction and evidence-span extraction from patient-provider secure messages. Across Llama models spanning 1B to 70B parameters, our prompt engineering improves zero-shot extraction, with an average label F1 gain of +14.46 and a +17.40 Span F1 gain. Building on supervised fine-tuning, STaR-DRO further improves accuracy and robustness, increasing average label F1 by +1.08/+2.20 and reducing average group-wise validation cross-entropy by 21.3%/14.8% relative to SFT and standard DRO, respectively. These findings significantly advance the reliability of automated communication mining for patient-centered clinical care analysis.
- 25 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSTaR-DRO: Stateful Tsallis Reweighting for Group-Robust Structured Prediction
Structured prediction with large language models requires more than accurate labels: outputs must satisfy ontology constraints, preserve valid structure, and ensure evidence-grounded decisions under label imbalance and heterogeneous group difficulty. To resolve these challenges, we present a unified framework for robust ontology-constrained generation. First, we introduce a modular prompt-engineering architecture combining XML-style structure, expert disambiguation rules, chain-of-thought reasoning, metadata-aware decision logic, schema contracts, and a self-validation gate. This targets recurrent in-context structured generation failures, including format drift, label ambiguity, evidence hallucination, and metadata-conditioned confusion. Second, we propose STaR-DRO, which combines Tsallis mirror ascent with sparse entmax-style primal mapback, EMA-smoothed group-loss tracking, rescaled ascent signals, and bounded excess-only multipliers. This upweights only persistently hard groups without suppressing easier ones, addressing conventional DRO’s dense Shannon-entropy exponentiated-gradient updates, high-variance stochastic reweighting, positive adversarial mass on groups that are not persistently hard, and performance costs from simplex competition. We evaluate the framework on EPPC Miner, a carefully chosen, clinically grounded representation of high-stakes structured prediction requiring hierarchical label prediction and evidence-span extraction from patient-provider secure messages. Across Llama models spanning 1B to 70B parameters, our prompt engineering improves zero-shot extraction, with an average label F1 gain of +14.46 and a +17.40 Span F1 gain. Building on supervised fine-tuning, STaR-DRO further improves accuracy and robustness, increasing average label F1 by +1.08/+2.20 and reducing average group-wise validation cross-entropy by 21.3%/14.8% relative to SFT and standard DRO, respectively. These findings significantly advance the reliability of automated communication mining for patient-centered clinical care analysis.
- Open Access
PVminer: A Domain-Specific Tool to Detect the Patient Voice in Patient Generated Data
Patient-generated text such as secure messages, surveys, and interviews contains rich expressions of the patient voice (PV), reflecting communicative behaviors and social determinants of health (SDoH). Traditional qualitative coding frameworks are labor intensive and do not scale to large volumes of patient-authored messages across health systems. Existing machine learning (ML) and natural language processing (NLP) approaches provide partial solutions but often treat patient-centered communication (PCC) and SDoH as separate tasks or rely on models not well suited to patient-facing language. We introduce PVminer, a domain-adapted NLP framework for structuring patient voice in secure patient-provider communication. PVminer formulates PV detection as a multi-label, multi-class prediction task integrating patient-specific BERT encoders (PV-BERT-base and PV-BERT-large), unsupervised topic modeling for thematic augmentation (PV-Topic-BERT), and fine-tuned classifiers for Code, Subcode, and Combo-level labels. Topic representations are incorporated during fine-tuning and inference to enrich semantic inputs. PVminer achieves strong performance across hierarchical tasks and outperforms biomedical and clinical pre-trained baselines, achieving F1 scores of 82.25% (Code), 80.14% (Subcode), and up to 77.87% (Combo). An ablation study further shows that author identity and topic-based augmentation each contribute meaningful gains. Pre-trained models, source code, and documentation will be publicly released, with annotated datasets available upon request for research use.
Open AccessPVminer: A Domain-Specific Tool to Detect the Patient Voice in Patient Generated Data
Patient-generated text such as secure messages, surveys, and interviews contains rich expressions of the patient voice (PV), reflecting communicative behaviors and social determinants of health (SDoH). Traditional qualitative coding frameworks are labor intensive and do not scale to large volumes of patient-authored messages across health systems. Existing machine learning (ML) and natural language processing (NLP) approaches provide partial solutions but often treat patient-centered communication (PCC) and SDoH as separate tasks or rely on models not well suited to patient-facing language. We introduce PVminer, a domain-adapted NLP framework for structuring patient voice in secure patient-provider communication. PVminer formulates PV detection as a multi-label, multi-class prediction task integrating patient-specific BERT encoders (PV-BERT-base and PV-BERT-large), unsupervised topic modeling for thematic augmentation (PV-Topic-BERT), and fine-tuned classifiers for Code, Subcode, and Combo-level labels. Topic representations are incorporated during fine-tuning and inference to enrich semantic inputs. PVminer achieves strong performance across hierarchical tasks and outperforms biomedical and clinical pre-trained baselines, achieving F1 scores of 82.25% (Code), 80.14% (Subcode), and up to 77.87% (Combo). An ablation study further shows that author identity and topic-based augmentation each contribute meaningful gains. Pre-trained models, source code, and documentation will be publicly released, with annotated datasets available upon request for research use.
- 1 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (12) , 5258-5277
Attention-Based Shape-Deformation Networks for Artifact-Free Geometry Reconstruction of Lumbar Spine From MR Images
Lumbar disc degeneration, a progressive structural wear and tear of lumbar intervertebral disc, is regarded as an essential role on low back pain, a significant global health concern. Automated lumbar spine geometry reconstruction from MR images will enable fast measurement of medical parameters to evaluate the lumbar status, in order to determine a suitable treatment. Existing image segmentation-based techniques often generate erroneous segments or unstructured point clouds, unsuitable for medical parameter measurement. In this work, we present UNet-DeformSA and TransDeformer: novel attention-based deep neural networks that reconstruct the geometry of the lumbar spine with high spatial accuracy and mesh correspondence across patients, and we also present a variant of TransDeformer for error estimation. Specially, we devise new attention modules with a new attention formula, which integrate tokenized image features and tokenized shape features to predict the displacements of the points on a shape template. The deformed template reveals the lumbar spine geometry in an image. Experiment results show that our networks generate artifact-free geometry outputs, and the variant of TransDeformer can predict the errors of a reconstructed geometry. Our code is available at https://github.com/linchenq/TransDeformer-Mesh.
- 1 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (12) , 5258-5277
Attention-Based Shape-Deformation Networks for Artifact-Free Geometry Reconstruction of Lumbar Spine From MR Images
Lumbar disc degeneration, a progressive structural wear and tear of lumbar intervertebral disc, is regarded as an essential role on low back pain, a significant global health concern. Automated lumbar spine geometry reconstruction from MR images will enable fast measurement of medical parameters to evaluate the lumbar status, in order to determine a suitable treatment. Existing image segmentation-based techniques often generate erroneous segments or unstructured point clouds, unsuitable for medical parameter measurement. In this work, we present UNet-DeformSA and TransDeformer: novel attention-based deep neural networks that reconstruct the geometry of the lumbar spine with high spatial accuracy and mesh correspondence across patients, and we also present a variant of TransDeformer for error estimation. Specially, we devise new attention modules with a new attention formula, which integrate tokenized image features and tokenized shape features to predict the displacements of the points on a shape template. The deformed template reveals the lumbar spine geometry in an image. Experiment results show that our networks generate artifact-free geometry outputs, and the variant of TransDeformer can predict the errors of a reconstructed geometry. Our code is available at https://github.com/linchenq/TransDeformer-Mesh.