More Like this

1. Saliency Maps of Images of Facial Disfigurements from Head and Neck Cancer

   Cook, M.; Bordes, M.C.; Lui, J.; Engelmann, J.; Chopra, D.; Reece, G.; Hanson, S.; Markey, M.K. (August 2019, 2019 BMES Conference Proceedings - REU Abstract Accepted Poster)

   Introduction: Head and neck cancer (HNC) and its treatment can result in facial disfigurement and functional defects in speech, swallowing, and vision that persist after reconstructive surgery. Body image concerns are pervasive among HNC patients, and a large portion of these concerns stem from worries about social interaction. Our overarching goal is to develop normative interventions to inform HNC patients about how others will respond to the changes in their facial appearance. In this study, we investigated saliency map algorithms for highlighting regions of interest on a clinically disfigured face that are expected to draw an observer’s eye based on color, intensity, etc. 

   more » « less
2. PRO-Based Stratification Improves Model Prediction for Toxicity and Survival of Head and Neck Cancer Patients

   https://doi.org/10.1109/JBHI.2024.3515092  

   Anyimadu, Eric A; Wang, Yaohua; Floricel, Carla; Kamel, Serageldin; Fuller, Clifton David; Zhang, Xinhua; Marai, G Elisabeta; Canahuate, Guadalupe M (February 2025, IEEE Journal of Biomedical and Health Informatics)
3. Improving Prediction of Late Symptoms using LSTM and Patient-reported Outcomes for Head and Neck Cancer Patients

   https://doi.org/10.1109/ICHI57859.2023.00047  

   Wang, Yaohua; Dijk, Lisanne Van; Mohamed, Abdallah_S R; Naser, Mohamed; Fuller, Clifton David; Zhang, Xinhua; Marai, G Elisabeta; Canahuate, Guadalupe (June 2023, IEEE Xplore)

   Patient-Reported Outcomes (PRO) are collected directly from the patients using symptom questionnaires. In the case of head and neck cancer patients, PRO surveys are recorded every week during treatment with each patient’s visit to the clinic and at different follow-up times after the treatment has concluded. PRO surveys can be very informative regarding the patient’s status and the effect of treatment on the patient’s quality of life (QoL). Processing PRO data is challenging for several reasons. First, missing data is frequent as patients might skip a question or a questionnaire altogether. Second, PROs are patient-dependent, a rating of 5 for one patient might be a rating of 10 for another patient. Finally, most patients experience severe symptoms during treatment which usually subside over time. However, for some patients, late toxicities persist negatively affecting the patient’s QoL. These long-term severe symptoms are hard to predict and are the focus of this study. In this work, we model PRO data collected from head and neck cancer patients treated at the MD Anderson Cancer Center using the MD Anderson Symptom Inventory (MDASI) questionnaire as time series. We impute missing values with a combination of K nearest neighbor (KNN) and Long Short-Term Memory (LSTM) neural networks, and finally, apply LSTM to predict late symptom severity 12 months after treatment. We compare performance against clinical and ARIMA models. We show that the LSTM model combined with KNN imputation is effective in predicting late-stage symptom ratings for occurrence and severity under the AUC and F1 score metrics. 

   more » « less
4. DITTO: A Visual Digital Twin for Interventions and Temporal Treatment Outcomes in Head and Neck Cancer

   https://doi.org/10.1109/TVCG.2024.3456160  

   Wentzel, Andrew; Attia, Serageldin; Zhang, Xinhua; Canahuate, Guadalupe; Fuller, Clifton David; Marai, G Elisabeta (January 2025, IEEE Transactions on Visualization and Computer Graphics)
5. A supervised machine learning model for identifying predictive factors for recommending head and neck cancer surgery

   https://doi.org/10.1002/hed.27674  

   Jiam, Max L; Xin, Kevin Z; Ha, Patrick K; Jiam, Nicole T (May 2024, Head & Neck)

   Abstract BackgroundNew patient referrals are often processed by practice coordinators with little‐to‐no medical background. Treatment delays due to incorrect referral processing, however, have detrimental consequences. Identifying variables that are associated with a higher likelihood of surgical oncological resection may improve patient referral processing and expedite the time to treatment. The study objective is to develop a supervised machine learning (ML) platform that identifies relevant variables associated with head and neck surgical resection. MethodsA retrospective cohort study was conducted on 64 222 patient datapoints from the SEER database. ResultsThe random forest ML model correctly classified patients who were offered head and neck surgery with an 81% accuracy rate. The sensitivity and specificity rates were 86% and 71%. The positive and negative predictive values were 85% and 73%. ConclusionsML modeling accurately predicts head and neck cancer surgery recommendations based on patient and cancer information from a large population‐based dataset. ML adjuncts for referral processing may decrease the time to treatment for patients with cancer. 

   more » « less