Maie St. John
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Maie St. John has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Deborah Jl Wong
Division of Hematology/Oncology, Department of Medicine, University of California Los Angeles (UCLA), Los Angeles, CA
1 shared publication - Eden R. Brauer
School of Nursing, University of California, Los Angeles, USA
1 shared publication - Eric M. Cheng
University of California, Los Angeles (UCLA), Los Angeles, USA
1 shared publication - Lauran K. Evans
Department of Head and Neck Surgery David Geffen School of Medicine at University of California Los Angeles California USA
1 shared publication - Patricia A. Ganz
David Geffen School of Medicine at UCLA, Los Angeles, CA
1 shared publication - Ramesh Shori
University of California, Los Angeles, David Geffen School of Medicine, Los Angeles, California, United States
1 shared publication
Latest Publications
- Open Access
- 21 July 2026
- Published by Springer Nature in Journal of Patient-Reported Outcomes
- Vol. 10 (1)
Using electronic patient reported outcomes (ePROs) to improve supportive care in ambulatory oncology care: a CFIR-guided pre-implementation formative evaluation
Background: Systematic collection of patient-reported outcomes via electronic health records can improve symptom assessment and supportive care delivery in oncology, yet implementation is often hindered by inadequate attention to end-user needs and organizational context. Guided by an implementation science framework, we conducted a formative evaluation to inform design and implementation planning of an electronic patient-reported outcome supportive care screening system in ambulatory oncology care, using head and neck cancer as a use case. Methods: A prospective mixed-methods observational study was conducted using semi-structured interviews as part of a pre-implementation formative evaluation. Eligible participants were target users of the proposed supportive care screening system, including (1) clinical staff and (2) patients with head and neck cancer from a single health care system. Using the Consolidated Framework for Implementation Research as a guiding framework, we elicited user feedback through interview and prototype engagement. Quantitative measures included patient symptom burden and validated pre-implementation measures of acceptability, appropriateness, feasibility, and usability. For data analyses, we used directed content analysis and matrix analysis for qualitative data, and descriptive and comparative statistics for quantitative assessments. Results: Forty-six target users (20 clinical staff, 26 patients) participated. Both groups expressed strong tension for change, recognizing many inadequacies of the current approach to symptom assessment. Clinical staff described current practices as “haphazard,” while patients recounted missed opportunities for support and preventable harm. Electronic health record integration enhanced perceived compatibility, though time constraints and alert fatigue emerged as potential barriers. Clinical staff emphasized the need for clear symptom thresholds and actionable referral criteria; patients emphasized that their sustained engagement depended on evidence of clinicians using the data in their care. Pre-implementation scores indicated strong readiness across both groups: acceptability (clinical staff 4.47/5, patients 4.21/5), appropriateness (4.46/5, 4.24/5), feasibility (4.41/5, 4.35/5), and usability (82.2/100, 85.6/100). Findings informed key design priorities and implementation strategies. Conclusions: Systematic pre-implementation formative evaluation revealed important barriers, facilitators, and contextual insights that shaped both intervention design and implementation approach. High scores on leading indicators suggest strong organizational readiness and perceived value. This user-informed development process provides a replicable model for optimizing electronic patient-reported outcome systems before deployment.
- 21 July 2026
- Published by Springer Nature in Journal of Patient-Reported Outcomes
- Vol. 10 (1)
Open AccessUsing electronic patient reported outcomes (ePROs) to improve supportive care in ambulatory oncology care: a CFIR-guided pre-implementation formative evaluation
Background: Systematic collection of patient-reported outcomes via electronic health records can improve symptom assessment and supportive care delivery in oncology, yet implementation is often hindered by inadequate attention to end-user needs and organizational context. Guided by an implementation science framework, we conducted a formative evaluation to inform design and implementation planning of an electronic patient-reported outcome supportive care screening system in ambulatory oncology care, using head and neck cancer as a use case. Methods: A prospective mixed-methods observational study was conducted using semi-structured interviews as part of a pre-implementation formative evaluation. Eligible participants were target users of the proposed supportive care screening system, including (1) clinical staff and (2) patients with head and neck cancer from a single health care system. Using the Consolidated Framework for Implementation Research as a guiding framework, we elicited user feedback through interview and prototype engagement. Quantitative measures included patient symptom burden and validated pre-implementation measures of acceptability, appropriateness, feasibility, and usability. For data analyses, we used directed content analysis and matrix analysis for qualitative data, and descriptive and comparative statistics for quantitative assessments. Results: Forty-six target users (20 clinical staff, 26 patients) participated. Both groups expressed strong tension for change, recognizing many inadequacies of the current approach to symptom assessment. Clinical staff described current practices as “haphazard,” while patients recounted missed opportunities for support and preventable harm. Electronic health record integration enhanced perceived compatibility, though time constraints and alert fatigue emerged as potential barriers. Clinical staff emphasized the need for clear symptom thresholds and actionable referral criteria; patients emphasized that their sustained engagement depended on evidence of clinicians using the data in their care. Pre-implementation scores indicated strong readiness across both groups: acceptability (clinical staff 4.47/5, patients 4.21/5), appropriateness (4.46/5, 4.24/5), feasibility (4.41/5, 4.35/5), and usability (82.2/100, 85.6/100). Findings informed key design priorities and implementation strategies. Conclusions: Systematic pre-implementation formative evaluation revealed important barriers, facilitators, and contextual insights that shaped both intervention design and implementation approach. High scores on leading indicators suggest strong organizational readiness and perceived value. This user-informed development process provides a replicable model for optimizing electronic patient-reported outcome systems before deployment.
- Open Access
- 1 January 2026
- Published by SPIE-Intl Soc Optical Eng in Biophotonics Discovery
- Vol. 3 (01) , 1-13
Machine-learning-based tumor segmentation and classification using dynamic optical contrast imaging for thyroid cancer
SignificanceThyroid cancer is the most common endocrine malignancy, and diagnosis is often challenging due to overlapping features between benign and malignant nodules. Fine-needle aspiration, the clinical gold standard, frequently yields indeterminate results and lacks spatial context, leading to unnecessary surgeries. Real-time margin assessment also remains limited. There is a need for accurate, label-free, spatially resolved imaging. Dynamic optical contrast imaging (DOCI), which measures autofluorescence lifetimes of endogenous fluorophores, offers a promising platform for intraoperative cancer detection.AimWe develop and evaluate a machine learning integrated DOCI framework to classify thyroid tissue subtypes and segment cancerous regions from ex vivo hyperspectral sections, with potential for real-time surgical use.ApproachFresh ex vivo thyroid specimens were imaged using a 23-channel DOCI acquisition. A pixel-level principal component analysis (PCA) and logistic regression classifier produced tissue probabilities, aggregated by a regional majority-vote gate to categorize specimens as normal, follicular, or papillary. Tumor-specific squeeze-and-excitation U-Net models were trained on voxel-only inputs for semantic segmentation. PCA-guided channel ablation identified a reduced spectral subset, and the pipeline was retrained using a compact 12-channel input.ResultsThe first two PCA components explained over 70% of spectral variance and yielded well-separated tissue clusters. The regional PCA classifier achieved 92.3% validation accuracy and 100% accuracy on the test set. Full 23-channel U-Net models delivered strong segmentation (papillary: Dice nonempty 0.829, balanced Dice 0.914; follicular: Dice nonempty 0.618, balanced Dice 0.809). Reduced-channel models preserved most performance and improved follicular segmentation (Dice nonempty 0.762), confirming spectral redundancy.ConclusionsIntegrating DOCI with interpretable machine learning enables accurate, label-free differentiation and segmentation of thyroid tissues. Channel reduction demonstrates that high performance is achievable with a compact spectral subset, supporting faster, more cost-efficient DOCI systems and future real-time intraoperative deployment.
- 1 January 2026
- Published by SPIE-Intl Soc Optical Eng in Biophotonics Discovery
- Vol. 3 (01) , 1-13
Open AccessMachine-learning-based tumor segmentation and classification using dynamic optical contrast imaging for thyroid cancer
SignificanceThyroid cancer is the most common endocrine malignancy, and diagnosis is often challenging due to overlapping features between benign and malignant nodules. Fine-needle aspiration, the clinical gold standard, frequently yields indeterminate results and lacks spatial context, leading to unnecessary surgeries. Real-time margin assessment also remains limited. There is a need for accurate, label-free, spatially resolved imaging. Dynamic optical contrast imaging (DOCI), which measures autofluorescence lifetimes of endogenous fluorophores, offers a promising platform for intraoperative cancer detection.AimWe develop and evaluate a machine learning integrated DOCI framework to classify thyroid tissue subtypes and segment cancerous regions from ex vivo hyperspectral sections, with potential for real-time surgical use.ApproachFresh ex vivo thyroid specimens were imaged using a 23-channel DOCI acquisition. A pixel-level principal component analysis (PCA) and logistic regression classifier produced tissue probabilities, aggregated by a regional majority-vote gate to categorize specimens as normal, follicular, or papillary. Tumor-specific squeeze-and-excitation U-Net models were trained on voxel-only inputs for semantic segmentation. PCA-guided channel ablation identified a reduced spectral subset, and the pipeline was retrained using a compact 12-channel input.ResultsThe first two PCA components explained over 70% of spectral variance and yielded well-separated tissue clusters. The regional PCA classifier achieved 92.3% validation accuracy and 100% accuracy on the test set. Full 23-channel U-Net models delivered strong segmentation (papillary: Dice nonempty 0.829, balanced Dice 0.914; follicular: Dice nonempty 0.618, balanced Dice 0.809). Reduced-channel models preserved most performance and improved follicular segmentation (Dice nonempty 0.762), confirming spectral redundancy.ConclusionsIntegrating DOCI with interpretable machine learning enables accurate, label-free differentiation and segmentation of thyroid tissues. Channel reduction demonstrates that high performance is achievable with a compact spectral subset, supporting faster, more cost-efficient DOCI systems and future real-time intraoperative deployment.
- Open Access
Cutaneous Squamous Cell Carcinoma in Immunocompromised Patients
Cutaneous squamous cell carcinoma (cSCC) is the second most common skin malignancy in the world, representing approximately 20% of all skin cancers. Immunosuppression is a well-established risk factor, contributing not only to the development of new cSCC lesions but also to more aggressive disease and increased mortality. Despite the National Comprehensive Cancer Network (NCCN) and the American Joint Committee on Cancer (AJCC) 8th edition updates recognizing immunosuppression as a risk factor for cSCC, standardized management protocols for these high-risk patients remain limited. As a result, treatment of this already high-risk group remains a significant challenge and highlights the need for dedicated research and attention to improve outcomes in this patient population. This review explores the current knowledge regarding cSCC in IS patients, outlines key gaps in the knowledge, and highlights recent clinical trials to further guide the evaluation and management of these patients.
Open AccessCutaneous Squamous Cell Carcinoma in Immunocompromised Patients
Cutaneous squamous cell carcinoma (cSCC) is the second most common skin malignancy in the world, representing approximately 20% of all skin cancers. Immunosuppression is a well-established risk factor, contributing not only to the development of new cSCC lesions but also to more aggressive disease and increased mortality. Despite the National Comprehensive Cancer Network (NCCN) and the American Joint Committee on Cancer (AJCC) 8th edition updates recognizing immunosuppression as a risk factor for cSCC, standardized management protocols for these high-risk patients remain limited. As a result, treatment of this already high-risk group remains a significant challenge and highlights the need for dedicated research and attention to improve outcomes in this patient population. This review explores the current knowledge regarding cSCC in IS patients, outlines key gaps in the knowledge, and highlights recent clinical trials to further guide the evaluation and management of these patients.