SSSatomi Shiraishi

Satomi Shiraishi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Satomi Shiraishi has a total of 96 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    VOICE-AE: automated CTCAE scoring from ambient clinical audio using speech recognition and large language models

    by Federico Mastroleo, Mariana Borras-Osorio, Jonathan E. Moonen, Jake A. Jordan, Keldon K. Lin, Qian Liu, Mi Zhou, Brad J. Stish, Daniel J. Ma, Robert W. Mutter, et al.

    Treatment-related adverse event documentation is labor-intensive and subject to inter-observer variability. We developed and evaluated VOICE-AE, a multi-stage pipeline integrating automatic speech recognition, speaker diarization and large language models (LLMs) for automated extraction and grading of Common Terminology Criteria for Adverse Events (CTCAE) toxicities from clinician-patient audio recordings. Encounters of prostate cancer patients undergoing radiotherapy were recorded and processed through VOICE-AE. Six blinded human raters independently graded 10 genitourinary/gastrointestinal toxicities from the same transcripts; hierarchical clustering identified an expert-rater subgroup that established the consensus reference standard. A total of 31 patients (39 encounters) were enrolled, yielding 390 encounter–toxicity assessments, of which 300 (76.9%) carried a reference grade of 0. Six-rater Krippendorff alpha was 0.677 for ordinal grading. Against reference standard, VOICE-AE achieved a mean weighted kappa of 0.717 and Krippendorff alpha of 0.669. Exact CTCAE grade match was 82.6% and within-one-grade accuracy 99.2%; because grade 0 predominated, a trivial classifier always predicting grade 0 would achieve 76.9% and 93.6% respectively. Performance was weaker at the grades that drive clinical action (grade 1 F1 0.58; grade 2 F1 0.57), and binary detection sensitivity was 77.8% with specificity 89.0% and precision 68.0%. In a like-for-like comparison restricted to the three expert-cluster raters, mean weighted kappa between VOICE-AE and those experts (0.666) fell just below agreement among the experts themselves (0.681). VOICE-AE performed within the range of inter-expert agreement, supporting the feasibility of automated, post-encounter AE documentation. Accuracy at grade ≥2 and detection of severe toxicity remain unproven.

  • Open Access

    Patient-reported outcomes in radiotherapy (PRO-RT) utilizing an integrated patient portal; results from a prospective single-arm trial

    by Yasamin Sharifzadeh, William S Harmsen, Jessica F Burlile, Kevin F Pyfferoen, Adam C Amundson, Mark R Waddle, Daniel J Ma, Satomi Shiraishi, Traci J Hammer, Nadia N Laack, et al.

    Objectives: This prospective single-arm trial assessed the use of an integrated patient platform (IPP) and effects on patient-reported outcomes survey (PROs) completion. Materials and Methods: Head and neck, breast, and thoracic cancer patients undergoing radiotherapy (RT) were approached for registration for an IPP for PRO administration and integration of multiple health portals. Results: Seventy-two patients were approached, 27 (38%) enrolled, and 18 used the IPP for PROs, 2 of which utilized integrative features. IPP user PRO completion rates were 100%, 89%, 89%, and 94% at baseline, end of RT, and 3- and 6-months post-RT, higher than the standard interface. Of the 45 who declined, 24 (53%) reported information security concerns. Discussion: While the IPP increased PRO completion, many declined enrollment, citing security concerns. Few utilized the portal’s integrative features. Conclusion: IPPs show promise in aiding PRO completion; continued efforts to demonstrate data safety and utility of IPPs are merited.

  • Large Language Model and Natural Language Processing Approach to Identify Cancer Recurrence From Pathology Reports

    by Diego Bayona, Daniel K. Ebner, Lydia Ekama, Benjamin Kamdem Talom, Satomi Shiraishi, Mariana Borras-Osorio, Federico Mastroleo, Yuliia Lozko, Javad Namazi, Mark Waddle, et al.

    PURPOSE: Cancer recurrence is a critical outcome for patients and physicians. Retrospective cancer recurrence data can evaluate recurrence-directed treatment and generate novel interventions targeting recurrent cancers. However, large cancer databases do not provide recurrence-related information, stymying study at scale, and consequently require significant manual record review. Automated evaluation of records may allow for the rapid generation of easily analyzed data sets, accelerating the evaluation of recurrence altogether. METHODS: Patients treated with radiation therapy at one tertiary referral center from 2010 to 2018 with a verified cancer status (cancer recurrence v no cancer recurrence) were identified. Patients with recurrent disease were initially identified through manual record review, and the associated pathology report was collected. Google Automated Machine Learning with Natural Language Processing (AutoNLP) and Google Gemini 1.5 Pro were used to generate a model for binary classification, with comparison to the gold-standard manually developed data set. RESULTS: A total of 7,054 patients were identified. 3,431 (48.6%) were female, with a median age of 64 years. Head and neck (1,482, 21%), breast (1,480, 21%), upper GI (1,307, 18.5%), and lung/thorax (1,107, 15.7%) were the most common disease sites. Recurrence was verified for 1,546 patients (21.9%) using pathology reports, of which 1,249 positive cases were paired with 651 negative pathology reports for model development. Google Gemini 1.5 Pro consistently outperformed AutoNLP across all measurements of accuracy, generating a greater absolute difference in precision, recall, negative predictive value, and specificity, and a higher likelihood of correct classification at the individual level, rendering Gemini superior in recurrence status extraction. CONCLUSION: AutoNLP and Google Gemini 1.5 Pro are promising tools for identifying recurrence from pathology reports, with the latter demonstrating superior overall performance, making it particularly suitable for clinical translation.

  • Large language models for toxicity extraction in oncology trials: A real-world benchmark in prostate radiotherapy

    by Federico Mastroleo, Mariana Borras-Osorio, Shiv P. Patel, Sarah Peterson, Renthony Wilson, Mi Zhou, Satomi Shiraishi, Andrew Y.K. Foong, David M. Routman, Mark R. Waddle


  • Open Access

    Deep Learning Driven Field Dose Prediction for Head and Neck Cancer Treated with Spot Scanning Proton Therapy

    by Brandon Reber, Satomi Shiraishi, Andrew Y. K. Foong, David Routman, Jing Qian

    Purpose: Accurate dose prediction is essential for automating radiotherapy planning. In spot scanning proton therapy (SSPT), dose evaluation is required at both the plan and field level. Evaluating individual treatment fields is critical to ensuring optimal beam angles are chosen to ensure target coverage and maximum organ-at-risk (OAR) sparing. Currently, however, no knowledge-based tools exist for predicting field-level doses for head and neck cancer (HNC) treated with SSPT. In this work, we aim to develop the first deep learning–based dose prediction model capable of field-level dose prediction for HNC treated with SSPT. Methods: A cohort of 62 HNC patients treated with SSPT was compiled for model development and evaluation. Collected patient data included treatment planning CTs, OAR masks, signed distance maps (SDMs), generated beam masks, and dose distributions. An encoder-decoder architecture enhanced with a cross-attention transformer bottleneck was used as the field prediction model. Comparison and ablation studies evaluated the model’s performance and determined the benefits of individual model components. Evaluation imaging metrics included mean absolute error, structural similarity index measure, and peak signal-to-noise ratio. Clinical performance was evaluated using dose-volume histogram metrics. Results: The best performing model from the ablation study was the full model using OAR masks, SDMs, generated beam masks and four-field dose prediction. The model outperformed the Distance Guided Dose Prediction (DGDP) and DeepLabV3 comparison models. The DGDP and DeepLabV3 comparison models had a mean validation set MAE performance of 1.268 Gy and 1.325 Gy, respectively, compared to our model’s mean validation set MAE performance of 0.949 Gy. The model’s final mean test set performance was MAE 1.024 Gy, SSIM 0.913, and PSNR 28.495 dB. Conclusions: We developed a cross-attention transformer–enhanced deep learning model that accurately predicts per-field dose for HNC treated with SSPT, demonstrating superior performance over state-of-the-art models limited to plan-level dose prediction.


  • Comparison of partial breast radiation modalities in women with early-stage breast cancer: a target trial emulation

    by Dean A Shumway, Kimberly S Corbin, Satomi Shiraishi, Magdoleen H Farah, Farah Fleti, Bashar Hasan, Mohammed M Firwana, Samer Saadi, Tarek Nayfeh, Moustafa Hegazi, et al.

    Background: Partial breast irradiation (PBI) is an effective treatment for early-stage breast cancer. However, evidence comparing different PBI modalities is limited. Methods: This trial emulation uses electronic health records from a multi-state large tertiary health system. Three PBI modalities were compared: proton radiation therapy, photon radiation therapy, and applicator-based brachytherapy. Inverse probability weighting (IPW) was used to balance confounders. The primary outcome was ipsilateral breast recurrence (IBR). Results: Between 2013 and 2023, 1041 women with early-stage breast cancer were treated with PBI. Of these, 1026 patients were included in the analyses. The median age was 66 years; 22.51% had ductal carcinoma in situ (DCIS), and 97.56% were estrogen receptor positive. With median follow-up of 38.2 months, the rate of IBR in the overall group was 3.05% (95% confidence interval [CI] = 1.99% to 4.65%) at 3 years. After IPW, compared with photon PBI, the hazard ratio for IBR for proton PBI was 2.11 (95% CI = 0.70 to 6.33) and for brachytherapy, 3.86 (95% CI = 1.39 to 10.69). Compared with proton PBI, the hazard ratio for IBR for brachytherapy was 1.83 (95% CI = 0.88 to 3.81). IBR risk was similar across PBI modalities for patients with tumor size ≤ 10 mm and in patients treated for DCIS. Among patients with tumor size >10 mm, the hazard ratio for IBR for brachytherapy was 7.64 (95% CI = 1.64 to 35.58) and 3.59 (95% CI = 1.22 to 10.56) relative to photons and protons, respectively. Conclusion: PBI with applicator-based brachytherapy was associated with higher IBR than photon or proton PBI in patients with tumor size > 10 mm, suggesting the need for more careful PBI patient selection for this modality.

  • Open Access

    Intensity-modulated proton radiotherapy spares musculoskeletal structures in regional nodal irradiation for breast cancer: a dosimetric comparison

    by Jessica F. Burlile, Satomi Shiraishi, Heather J. Gunn, Jennifer L. Bradt, Haley M. Kroeplin, Karen G. Lang, Jenna K. Cimmiyotti, Nicolas Depauw, Connie Y. Chang, Kevin M. Brom, et al.

    Background and purpose: Regional nodal irradiation (RNI) for breast cancer delivers radiation in proximity to the shoulder and torso, and radiation exposure may contribute to long-term upper extremity and postural morbidity. To date, no studies have assessed the differential dosimetric impact of proton versus photon radiation on shoulder and torso anatomy. This study examined clinically relevant musculoskeletal (MSK) structures and assessed the dose delivered with each modality. Patients/material and methods: Ten MSK structures were contoured on IMPT (intensity-modulated proton therapy) and VMAT (volumetric modulated arc therapy) plans for 30 patients receiving RNI. Relevant dose metrics were compared for each of the structures. Intensity-modulated proton therapy dose was calculated using the relative biological effective value of 1.1. Hypo-fractionated plans were scaled to the equivalent dose in 2 Gy fractions (EQD2) using an alpha/beta ratio of four. Wilcoxon signed rank sum tests compared doses. Select three-dimensional and optimised VMAT plans were also informally compared. Results and interpretation: Each of the 10 structures received a statistically significantly lower dose with the use of IMPT compared with VMAT. Differences were greatest for posterior structures, including the trapezius, latissimus dorsi and glenohumeral joint. Mean absolute differences were as great as 23 Gy (supraspinatus D5cc) and up to 30-fold dose reductions were observed (deltoid D50cc). An average 3.7-fold relative dose reduction existed across all structures. Measures of low/intermediate dose (V15Gy and D50cc) showed the largest differences. Intensity-modulated proton therapy results in statistically lower radiation exposure to relevant shoulder and torso anatomy compared to photon radiation for patients requiring RNI. Prospective study is needed to correlate functional outcomes with radiation dose.


  • Assessment of minimum target dose as a predictor of local failure after spine SBRT

    by Roman O. Kowalchuk, Trey C. Mullikin, Grant M. Spears, Benjamin A. Johnson-Tesch, Peter S. Rose, Brittany L. Siontis, Dong Kun Kim, Brian A. Costello, Jonathan M. Morris, Robert W. Gao, et al.


  • Open Access

    Oncologic Outcomes for Head and Neck Skin Malignancies Treated with Protons

    by Jidapa Bridhikitti, Jason K. Viehman, W. Scott Harmsen, Adam C. Amundson, Satomi Shiraishi, Daniel W. Mundy, Jean-Claude M. Rwigema, Lisa A. McGee, Samir H. Patel, David M. Routman, et al.

    Radiation therapy (RT) is the standard treatment for patients with inoperable skin malignancies of the head and neck region (H&N), and as adjuvant treatment post surgery in patients at high risk for local or regional recurrence. This study reports clinical outcomes of intensity-modulated proton therapy (IMPT) for these malignancies. We retrospectively reviewed cases involving 47 patients with H&N malignancies of the skin (squamous cell, basal cell, melanoma, Merkel cell, angiosarcoma, other) who underwent IMPT for curative intent between July 2016 and July 2019. Overall survival was estimated via Kaplan-Meier analysis, and oncologic outcomes were reported as cumulative incidence with death as a competing risk. The 2-year estimated local recurrence rate, regional recurrence rate, local regional recurrence rate, distant metastasis rate, and overall survival were 11.1% (95% confidence interval [CI], 4.1%-30.3%), 4.4% (95% CI, 1.1%-17.4%), 15.5% (95% CI, 7%-34.3%), 23.4% (95% CI, 5.8%-95.5%), and 87.2% (95% CI, 75.7%-100%), respectively. No patient was reported to have a grade 3 or higher adverse event during the last week of treatment or at the 3-month follow-up visit. IMPT is safe and effective in the treatment of skin malignancies of the H&N.