Peng CM, Chen CW, Hsieh CH, Cheng YY, Liao CH, Hsieh MF, Lin SC, Liu MC, Liu YJ. Radiomics meets sarcopenia: Machine learning-based multimodal modeling for esophageal cancer outcomes. World J Gastrointest Oncol 2025; 17(10): 111399 [PMID: 41114100 DOI: 10.4251/wjgo.v17.i10.111399]
Corresponding Author of This Article
Yi-Jui Liu, PhD, Professor, Department of Automatic Control Engineering, Feng Chia University, No. 100 Wenhwa Road, Taichung 407, Taiwan. erliu@fcu.edu.tw
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Minireviews
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Co-first authors: Cheng-Ming Peng and Chun-Wen Chen.
Co-corresponding authors: Ming-Cheng Liu and Yi-Jui Liu.
Author contributions: Peng CM, Chen CW, Hsieh CH, Cheng YY, Liao CH, Hsieh MF, and Liu MC provided expertise and experience in esophageal cancer and sarcopenia; Lin SC and Liu YJ contributed expertise in radiomics and machine learning model; Peng CM and Liu YJ conducted a survey and reviewed relevant studies; Liu MC and Liu YJ drafted the manuscript; Peng CM, Chen CW, Liu MC, and Liu YJ revised the manuscript; All authors have reviewed and approved the final version of the manuscript. Peng CM and Chen CW contributed equally to this manuscript and are co-first authors. Liu MC and Liu YJ contributed equally to this work as co-corresponding authors.
Supported by Feng Chia University/Chung Shan Medical University, No. FCU/CSMU 112-001; and Taiwan National Science and Technology Council, No. 111-2314-B-035-001-MY3.
Conflict-of-interest statement: The authors have no conflicts of interest to declare.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Yi-Jui Liu, PhD, Professor, Department of Automatic Control Engineering, Feng Chia University, No. 100 Wenhwa Road, Taichung 407, Taiwan. erliu@fcu.edu.tw
Received: June 30, 2025 Revised: July 15, 2025 Accepted: September 2, 2025 Published online: October 15, 2025 Processing time: 106 Days and 22.3 Hours
Abstract
Esophageal cancer is a highly aggressive malignancy often diagnosed at an advanced stage, with poor prognosis and high recurrence rates despite curative treatment. Accurate prognostic tools are urgently needed to guide personalized management strategies. Recent research has demonstrated significant potential of integrating quantitative imaging biomarkers, specifically radiomics and sarcopenia, with machine learning (ML) techniques to enhance outcome prediction. This review systematically summarizes six recent studies (2022-2024) exploring integrated ML models combining sarcopenia and radiomics biomarkers with clinical parameters to predict survival in patients with esophageal and gastroesophageal cancers. Sample sizes ranged from 83 to 243 patients, with studies utilizing various imaging modalities (positron emission tomography/computed tomography and computed tomography) and model analysis approaches, including Cox regression, random forest, and light gradient boosting machine. These models incorporated features such as skeletal muscle indices, tumor texture, and shape descriptors. Models that combined clinical data, radiomics, and sarcopenia outperformed those using single modalities. These findings support the utility of multimodal imaging biomarkers in developing robust, individualized prognostic models. However, the retrospective nature of most studies highlights the need for standardization and external validation. This review underscores the potential of multimodal ML-based models in enhancing personalized risk stratification and treatment planning for esophageal cancer.
Core Tip: This review highlights recent advances in machine learning models that integrate radiomics and sarcopenia biomarkers for outcome prediction in esophageal and gastroesophageal cancers. Multimodal models consistently outperform single-feature models, offering more accurate and personalized prognostic assessments. The integration of radiomics-derived tumor features and sarcopenia-related body composition indices offers deeper insights into tumor biology and patient health status. However, achieving clinical translation requires addressing methodological variability and ensuring rigorous validation. These advanced imaging analytics hold significant promise for personalized patient management in esophageal cancer.
Citation: Peng CM, Chen CW, Hsieh CH, Cheng YY, Liao CH, Hsieh MF, Lin SC, Liu MC, Liu YJ. Radiomics meets sarcopenia: Machine learning-based multimodal modeling for esophageal cancer outcomes. World J Gastrointest Oncol 2025; 17(10): 111399
Esophageal cancer is often diagnosed at an advanced stage due to the absence of obvious symptoms in its early course, contributing to its poor prognosis[1]. Moreover, the disease encompasses diverse histological subtypes, each associated with distinct prognostic implications[2,3]. Because of delayed diagnosis and aggressive tumor progression, esophageal cancer ranks as the seventh most common cancer worldwide and the sixth leading cause of cancer-related mortality[4]. Despite advances in treatment, high rates of postoperative recurrence and a persistently low 5-year survival rate continue to present major clinical challenges[5,6]. Consequently, the management of esophageal cancer remains difficult and demands ongoing improvement in early detection and personalized treatment strategies.
Sarcopenia, originally defined as the unintentional loss of skeletal muscle mass and strength that occurs with age[7], has recently been widely recognized as a distinct disease and an important risk factor for functional impairment, physical disability, reduced quality of life, and increased mortality[8]. The progression to sarcopenia is multifactorial, often driven by dysphagia, reduced nutrient intake, and the elevated metabolic demands associated with aggressive tumor biology, all of which contribute to systemic inflammation and subsequent muscle wasting[9]. The severity of the systemic inflammatory response and the risk of developing cancer-related cachexia are influenced by various factors, with tumor type being one of the most critical determinants[10]. Sarcopenia has been widely reported as a negative prognostic factor in multiple cancers, particularly those with high prevalence such as pancreatic, esophageal, gastrointestinal, head and neck, and lung cancers[11].
Medical imaging modalities such as ultrasound, computed tomography (CT), magnetic resonance imaging, and positron emission tomography (PET) play a crucial role in cancer diagnosis and management. The key characteristic of tumor biology, high intratumoral heterogeneity, can be effectively visualized through these imaging techniques[12]. Radiomics, an emerging field, enables the automated high-throughput extraction of large volumes of quantitative imaging features, offering a novel approach to analyzing medical images[13]. Recent studies have demonstrated that these high-dimensional radiomic features are valuable in addressing a range of clinical challenges across cancer care continuum, including diagnosis, treatment response assessment, and prognosis prediction[14].
Machine learning (ML) has become widely utilized in oncology for cancer diagnosis, treatment planning, and outcome prediction, although its clinical implementation remains challenging[15]. ML algorithms excel at processing and interpreting complex, heterogeneous, and voluminous datasets, which are characteristic of oncology data derived from multiple modalities. Both sarcopenia and radiomics provide clinically relevant information related to cancer treatment outcomes. Consequently, an increasing number of studies are investigating integrated models that combine sarcopenia and radiomics with ML model to predict survival outcomes in esophageal cancer patients. This review synthesizes recent advances in multimodal ML prognostic models integrating radiomics and sarcopenia biomarkers for survival outcome estimation in esophageal cancer. Specifically, we evaluate model performance, explore their potential clinical application in decision-making for treatment strategies, and discuss current limitations while proposing future research directions and clinical translation.
LITERATURE SEARCH
We used the PubMed medical database for articles published search. The search terms were implemented using the following Boolean search algorithm: (“sarcopenia” OR “muscle loss”) AND (“esophageal cancer” OR “esophageal tumor” OR “oesophageal cancer” OR “oesophageal tumor”) AND (“radiomic” OR “radiomics”). A literature review was conducted to identify studies that investigated prognostic outcome models involving sarcopenia, radiomics, and their combination. A total of 94968 studies were retrieved by the Boolean search algorithm terms (“sarcopenia” OR “muscle loss”); 89920 studies were retrieved using the terms (“esophageal cancer” OR “esophageal tumor” OR “oesophageal cancer” OR “oesophageal tumor”); 15317 studies were retrieved using the terms (“radiomic” OR “radiomics”). Only seven articles remained after the intersection of all three search groups. Finally, six original research articles were included in the review after removing one commentary article related to an included original study[16-21].
From 2022 to 2024, six studies have primarily focused on integrating sarcopenia and radiomics to predict clinical outcomes in esophageal cancers, as summarized in Table 1[16-21]. Tumor types included esophageal squamous cell carcinoma, esophagogastric adenocarcinoma, and gastroesophageal cancer. All studies utilized pretreatment imaging, with one study additionally incorporating follow-up scans[18]. The prognostic targets varied and included overall survival (OS)[16,17,19-21], progression-free survival (PFS), metastatic disease, relapse-free survival (RFS), and progressive disease (PD) and sarcopenia[16-18,20,21].
Table 1 Six studies published between 2022 and 2024 combined sarcopenia and radiomics assessments to predict clinical outcomes in esophageal cancer.
Body composition quantification was consistently conducted using CT-derived indices of skeletal muscle and adipose tissue at the third lumbar vertebral level. Most studies used non-contrast-enhanced CT, except Vogele et al[18], which employed intravenous contrast. Radiomic features were typically extracted from tumor regions of interest (ROIs) on PET and CT scans[16,17,20,21]. In contrast, two studies analyzed ROIs in muscle groups such as the psoas, quadratus lumborum, erector spinae, and abdominal wall muscles[18,19]. This body of literature reflects a growing trend toward integrating radiomic features and quantitative body composition data into survival prediction models. Collectively, these studies explore the prognostic value of combining tumor-specific imaging features with systemic physiological indicators, supporting a more holistic approach to risk stratification in patients with esophageal and gastroesophageal cancers.
COLLECTED FEATURES FROM CLINICAL, SARCOPENIA, AND RADIOMIC DATA
The feature sets comprising clinical data, body composition indices, and radiomics, along with their corresponding predictive models, are comprehensively summarized for the six studies in Table 2. Most studies utilized Cox regression, with or without nomogram construction[16,17,19-21], whereas others incorporated ML techniques, including light gradient boosting machine, decision trees, random forest (RF), and k-nearest neighbors[17,18]. Clinical variables frequently analyzed included patient demographics (e.g., age, sex, race), body mass index, Eastern Cooperative Oncology Group performance status, tumor, node, and metastasis stage, tumor histology, and blood-derived biomarkers such as C-reactive protein, albumin, neutrophil-lymphocyte ratio, platelet-lymphocyte ratio, prognostic nutritional index, and modified Glasgow prognostic score. Body composition quantification primarily relied on skeletal muscle index (SMI) at the third lumbar vertebral level[16,17,19-21], with additional metrics including the total psoas area, total psoas index (TPI), visceral adipose tissue index, subcutaneous adipose tissue index, and psoas muscle index[16,18,19].
Table 2 Summary of the predictive models and collected features, including clinical data, body composition indices, and radiomics, used in the six reviewed studies.
Sarcopenia was defined using either sex-specific SMI thresholds (< 34.4 cm2/m2 for females, < 45.4 cm2/m2 for males) or the threshold of the lowest quartile of TPI values[16-18,20,21]. Radiomics analyses involved high-throughput extraction of quantitative features ranging from 42 to 837, derived from PET/CT-based tumor or skeletal muscle ROIs. Feature selection methods varied across studies, including least absolute shrinkage and selection operator regression, variance inflation factor analysis[19], and significance testing between clinical groups[16-19,21]. This integrated modeling approach has been consistently investigated to determine whether it enhances prognostic accuracy compared to single-modality models, and to evaluate its utility in supporting individualized risk stratification and treatment planning in patients with esophageal cancers.
EVALUATION OF PROGNOSTIC MODELS IN ESOPHAGEAL CANCER
A detailed review of six studies in the sample size, selected features, the final or best models, and performances as summarized in Table 3, underscores a growing consensus that integrating radiomics features and sarcopenia-related body composition metrics with conventional clinical data significantly improves prognostic accuracy for esophageal and gastroesophageal cancers. Sample sizes across the reviewed studies ranged from 83 to 243 patients. These studies evaluated diverse patient populations using different imaging modalities. Four studies employed PET/CT-based analysis[16,17,20,21], while two relied exclusively on CT imaging[18,19]. Modeling techniques varied, including Cox proportional hazards models[16,17,19-21], nomogram-based visualization[16,19,21], and ML approaches such as RF and light gradient boosting machine[17,18]. Despite methodological variations, all studies consistently reported that models integrating clinical, body composition, and radiomic parameters improved the predictive accuracy for clinical outcomes, including OS, PFS, RFS, PD, and metastatic disease status[16-19,20,21].
Table 3 Summary of selected features, final or best-performing models, and predictive performance in the six reviewed studies.
Across these reviewed studies, there was a consistent emphasis on integrating clinical parameters, body composition indices, and radiomics into the prognostic modeling for esophageal and gastroesophageal cancers. Selected clinical features included tumor staging, Eastern Cooperative Oncology Group performance status, lymph nodal involvement, and demographic factors such as age, sex, and race[16-19,20,21]. Additional variables included body mass index, chemotherapy status, and blood-based biomarkers such as albumin, prognostic nutritional index, platelet-lymphocyte ratio, and C-reactive protein/albumin ratio[16,19].
Selected body composition features
Sarcopenia-related measurements were consistently included across studies, with SMI being the most commonly used index[16,17,20]. Other studies incorporated broader body composition markers such as the visceral adipose tissue index and direct sarcopenia classification to enrich the prognostic models[16,21].
Selected radiomic features
Radiomics features were widely employed across all reviewed studies, with notable variation in both source and type. Several studies utilized tumor intensity-based metrics and histogram-based features extracted from PET/CT images[20,21]. Shape- and texture-based radiomic features - such as shape, gray-level co-occurrence matrix, gray-level run-length matrix, gray-level zone length matrix, and neighborhood gray-level difference matrix - were frequently analyzed in tumor lesions from PET/CT imaging[16,17,20]. Zhou et al[16] further derived radiomic scores (Rad-scores) to quantitatively summarize their predictive potential. In contrast, studies by Vogele et al[18] and Iwashita et al[19] focused on CT-derived features from skeletal muscle regions, including histogram metrics (e.g., histogram P10th, histogram robust mean absolute deviation, interquartile range) and advanced texture features obtained through wavelet transformations[18,19]. This methodological variability highlights the diverse yet complementary roles of radiomic analyses in enhancing prognostic assessments across all six studies.
Performance in predicting outcomes
The performance outcomes, summarized in Table 3, reinforce the value of integrating radiomics features and sarcopenia metrics with clinical data enhance prognostic accuracy in patients with esophageal or gastroesophageal cancer. Zhou et al[16] demonstrated robust predictive capability for both PFS and OS with a combined model incorporating clinical parameters, body composition indices, and radiomics features, achieving concordance index (C-index)/area under the curve (AUC) values of 0.810 and 0.806, respectively. Hinzpeter et al[17] reported that their integrated model combining radiomics, clinical data, and SMI attained an 80% accuracy and an AUC of 0.88 for predicting metastatic disease. Although specific C-index values were not reported for OS, Kaplan-Meier survival curves confirmed the statistical significance of the selected predictors. Vogele et al[18] showed excellent classification performance in predicting sarcopenia (accuracy: 0.90 ± 0.03; AUC: 0.96 ± 0.02), PD (accuracy: 0.88 ± 0.04; AUC: 0.93 ± 0.04), and PD with sarcopenia (accuracy: 0.93 ± 0.04; AUC: 0.97 ± 0.04), highlighting the utility of CT-based muscle radiomics with an RF model. Iwashita et al[19] validated a prognostic model combining clinical and radiomic features for OS prediction, yielding 75% accuracy, 92% specificity, 75% sensitivity, and a C-index of 0.88. Hinzpeter et al[20] also found that models integrating PET/CT radiomics and sarcopenia outperformed those using clinical data along or combining only one of the modalities, achieving a 24-month OS AUC of 0.88. Similarly, Anconina et al[21] found that incorporating SMI and CT-derived features alongside clinical data improved predictive performance for both RFS and OS, with AUCs of 0.80 and 0.81 at 24 months and 30 months, respectively.
DISCUSSION
Current prognosis and the need for accurate prediction
Despite advances in surgical techniques and postoperative management, including concurrent chemoradiotherapy (CCRT), esophageal cancer remains a leading cause of cancer-related morbidity and mortality worldwide. The prognosis for esophageal cancer continues to be poor, with reported 3-year and 5-year OS rates in China of approximately 49.98% and 39.07%, respectively[22]. Globally, the 5-year survival rate for both esophageal adenocarcinoma and esophageal squamous cell carcinoma across all stages remains below 25%[23]. These poor outcomes underscore the urgent need for accurate prognostic tools to enable more precise patient stratification and to support personalized treatment strategies that can ultimately improve clinical outcomes.
Clinical benefits of prognostic prediction in treatment strategies
Developing accurate outcome prediction models based on medical imaging holds significant potential for improving esophageal cancer management. First, predicting treatment response and prognosis facilitates individualized strategies for CCRT and plays a critical role in optimizing therapeutic decisions and enhancing patient outcomes[23,24]. For instance, forecasting survival duration and treatment efficacy prior to initiating triple therapy in advanced adenocarcinoma of the esophagogastric junction enables effective risk stratification, helping clinicians classify patients into high-risk, medium-risk, and low-risk groups for OS and PFS[25]. Second, neoadjuvant therapy (NAT) followed by surgery is the standard of care for patients with locally advanced esophageal or gastroesophageal junction cancer[26]. However, not all patients respond favorably to NAT. Early prediction of response to neoadjuvant chemotherapy can identify non-responders, helping to avoid unnecessary treatment delays and reduce the risk of adverse drug reactions[24]. Third, while the rate of achieving pathological complete response after neoadjuvant chemoradiotherapy remains relatively limited (the prevalence of pathological complete response 9%, 95% confidence interval: 6%-14% for squamous cell carcinomas)[27], patients who do attain pathological complete response may be suitable candidates for organ preservation and active surveillance strategies, offering meaningful clinical benefits[28].
Additionally, due to the often subtle or nonspecific symptoms in the early stages, esophageal and esophagogastric cancers are frequently diagnosed at locally advanced stages[29]. Even among patients receiving curative treatment, such as surgical resection or definitive chemoradiotherapy, a substantial proportion, up to 50%, experience disease recurrence following curative surgery[5,30] or definitive chemoradiotherapy[31,32]. Given the invasive nature and limited availability of tissue-based molecular biomarkers derived from biopsy or surgical specimens, there is growing interest in noninvasive imaging biomarkers. These may provide valuable insight for optimizing treatment sequencing, supporting individualized CCRT strategies, enabling early identification of complete or objective responses, and improving relapse risk prediction[23].
Prognostic value of radiomics and sarcopenia
In radiomics research, previous studies have demonstrated that radiomic analysis holds significant promise for advancing esophageal cancer management, particularly in diagnostic differentiation, tumor staging, treatment response prediction, and prognostication[23,24]. Regarding sarcopenia, numerous studies have shown that skeletal muscle wasting during NAT is a critical and independent prognostic factor associated with poor survival outcomes in patients with esophageal cancer[33,34]. Specifically, severe skeletal muscle wasting during NAT has been identified as a significant predictor of inferior RFS[33,34]. These findings underscore the prognostic value of a dynamic and potentially modifiable physiological parameter that may significantly influence treatment efficacy and overall patient outcomes.
Both radiomics and sarcopenia have individually demonstrated prognostic value in patients with esophageal cancer. Recently, several studies have integrated both biomarkers with clinical data to develop more robust outcome prediction models. In this review, six studies were examined that combined radiomic features and sarcopenia metrics for prognostic assessment in esophageal and gastroesophageal cancers[16-21]. These studies consistently underscore the growing importance of integrating multimodal imaging features with clinical and body composition data to enhance prognostic accuracy. They highlight radiomics and sarcopenia as valuable, complementary biomarkers for predicting patient outcomes. Collectively, these findings support a more holistic approach to outcome prediction, moving beyond traditional clinical staging to incorporate quantitative imaging analysis (radiomics) and body composition evaluation (sarcopenia).
Limitations of current studies
Retrospective study design, varied cohorts, and limited sample sizes: All six reviewed studies were retrospective single center investigations with relatively small sample sizes and diverse patient cohorts[16-21]. Sample sizes ranged from 83 to 243 patients. Three studies included fewer than 100 participants (83, 91, and 98), two enrolled 100-200 patients (128 and 145), and one study had 234 patients. The included cohorts varied in cancer type, encompassing esophageal cancer, gastroesophageal cancer, and gastric cancer. Tumor histology also varied across studies, including squamous cell carcinoma, adenocarcinoma, and one study that did not specify tumor subtype.
Varied definitions of sarcopenia and lack of imaging standardization: Different studies used inconsistent definitions for sarcopenia, such as sex-specific SMI cut-off values or the lowest quartile of TPI. This variability complicates direct comparisons across studies and limits generalizability. Additionally, variation in imaging acquisition parameters and post-processing protocols, especially in retrospective studies conducted over several years, can impact radiomics feature stability and reproducibility[35].
Lack of external validation: A key limitation across six studies was the absence of external validation cohorts for the developed models. While most studies employed internal cross-validation techniques to ensure statistical robustness, these methods do not fully address the issue of generalizability to broader and more diverse patient populations.
Future research needs
Although limitations such as retrospective study design and limited sample sizes, all six studies[16-21] consistently demonstrated that combined models consistently outperformed those based on single parameters or modalities. This reinforces the potential of these advanced methods in guiding personalized cancer care. However, to overcome the remaining challenges, future research in esophageal cancer could focus on the following directions.
Multicenter studies with external validation: Small sample sizes from single-institution studies inherently limit control over confounding variables and introduce potential biases. A common recommendation across the reviewed literature is the need for future research collaborations involving external institutions to validate findings in larger and more diverse cohorts. There is a need for prospective, harmonized investigations to ensure consistency in imaging protocols and enhance generalizability.
Standardization of image processing and sarcopenia assessment: To ensure reliable clinical integration, there is an urgent need for harmonization of imaging acquisition protocols, radiomics methodologies, feature selection processes, and the choice of ML models and workflows. For sarcopenia, future research should focus on establishing diagnostic cut-off points tailored to specific patient populations. Additionally, comparative evaluation of TPI and SMI as diagnostic and prognostic indices is warranted. Furthermore, incorporating muscle function assessments in prospective studies will help align with the full clinical definition of sarcopenia.
Addressing clinical translation challenges: To advance clinical translation, future studies should explore broader patient populations beyond specific histological types and disease stages and assess post-therapy radiomic changes to improve prediction of treatment response and outcomes. Ultimately, while current evidence supports the prognostic value of integrating PET/CT radiomics and sarcopenia with clinical data, robust, standardized, and externally validated multicenter studies are essential to achieve their full clinical impact.
CONCLUSION
Collectively, these findings emphasize that multimodal prognostic models, particularly those integrating radiomics and sarcopenia metrics, consistently outperform models based solely on clinical parameters or combinations of only two of the three index types. The consistent improvements in model discrimination and calibration across studies suggest that combining radiomics and body composition analysis with clinical data can more accurately reflect tumor biology and host response, offering a more nuanced and individualized approach to risk stratification. These results emphasize the importance of multidisciplinary integration in oncologic imaging research and reinforce the potential of advanced imaging analytics in guiding personalized treatment planning and follow-up strategies.
Footnotes
Provenance and peer review: Invited article; Externally peer reviewed.
Peer-review model: Single blind
Corresponding Author’s Membership in Professional Societies: International Society for Magnetic Resonance in Medicine, 44944.
Specialty type: Oncology
Country of origin: Taiwan
Peer-review report’s classification
Scientific Quality: Grade C, Grade C
Novelty: Grade D, Grade D
Creativity or Innovation: Grade C, Grade C
Scientific Significance: Grade C, Grade D
P-Reviewer: Tasci B, PhD, Associate Professor, Türkiye S-Editor: Zuo Q L-Editor: Filipodia P-Editor: Zhang XD
Cruz-Jentoft AJ, Baeyens JP, Bauer JM, Boirie Y, Cederholm T, Landi F, Martin FC, Michel JP, Rolland Y, Schneider SM, Topinková E, Vandewoude M, Zamboni M; European Working Group on Sarcopenia in Older People. Sarcopenia: European consensus on definition and diagnosis: Report of the European Working Group on Sarcopenia in Older People.Age Ageing. 2010;39:412-423.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 6987][Cited by in RCA: 8557][Article Influence: 570.5][Reference Citation Analysis (0)]
Hinzpeter R, Mirshahvalad SA, Kulanthaivelu R, Kohan A, Ortega C, Metser U, Liu A, Farag A, Elimova E, Wong RKS, Yeung J, Jang RW, Veit-Haibach P. Gastro-Esophageal Cancer: Can Radiomic Parameters from Baseline (18)F-FDG-PET/CT Predict the Development of Distant Metastatic Disease?Diagnostics (Basel). 2024;14:1205.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 1][Reference Citation Analysis (0)]
Vogele D, Mueller T, Wolf D, Otto S, Manoj S, Goetz M, Ettrich TJ, Beer M. Applicability of the CT Radiomics of Skeletal Muscle and Machine Learning for the Detection of Sarcopenia and Prognostic Assessment of Disease Progression in Patients with Gastric and Esophageal Tumors.Diagnostics (Basel). 2024;14:198.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 8][Reference Citation Analysis (0)]
Iwashita K, Kubota H, Nishioka R, Emoto Y, Kawahara D, Ishihara T, Miyawaki D, Nishibuchi I, Nagata Y, Sasaki R. Prognostic Value of Radiomics Analysis of Skeletal Muscle After Radical Irradiation of Esophageal Cancer.Anticancer Res. 2023;43:1749-1760.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 6][Reference Citation Analysis (0)]
Anconina R, Ortega C, Metser U, Liu ZA, Elimova E, Allen M, Darling GE, Wong R, Taylor K, Yeung J, Chen EX, Swallow CJ, Jang RW, Veit-Haibach P. Combined 18 F-FDG PET/CT Radiomics and Sarcopenia Score in Predicting Relapse-Free Survival and Overall Survival in Patients With Esophagogastric Cancer.Clin Nucl Med. 2022;47:684-691.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 7][Cited by in RCA: 25][Article Influence: 8.3][Reference Citation Analysis (0)]
Yang Z, Gong J, Li J, Sun H, Pan Y, Zhao L. The gap before real clinical application of imaging-based machine-learning and radiomic models for chemoradiation outcome prediction in esophageal cancer: a systematic review and meta-analysis.Int J Surg. 2023;109:2451-2466.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 17][Reference Citation Analysis (0)]
Kang YK, Yook JH, Park YK, Lee JS, Kim YW, Kim JY, Ryu MH, Rha SY, Chung IJ, Kim IH, Oh SC, Park YS, Son T, Jung MR, Heo MH, Kim HK, Park C, Yoo CH, Choi JH, Zang DY, Jang YJ, Sul JY, Kim JG, Kim BS, Beom SH, Cho SH, Ryu SW, Kook MC, Ryoo BY, Kim HK, Yoo MW, Lee NS, Lee SH, Kim G, Lee Y, Lee JH, Noh SH. PRODIGY: A Phase III Study of Neoadjuvant Docetaxel, Oxaliplatin, and S-1 Plus Surgery and Adjuvant S-1 Versus Surgery and Adjuvant S-1 for Resectable Advanced Gastric Cancer.J Clin Oncol. 2021;39:2903-2913.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 31][Cited by in RCA: 243][Article Influence: 60.8][Reference Citation Analysis (0)]
Xi M, Yang Y, Zhang L, Yang H, Merrell KW, Hallemeier CL, Shen RK, Haddock MG, Hofstetter WL, Maru DM, Ho L, Wu CC, Liu M, Lin SH. Multi-institutional Analysis of Recurrence and Survival After Neoadjuvant Chemoradiotherapy of Esophageal Cancer: Impact of Histology on Recurrence Patterns and Outcomes.Ann Surg. 2019;269:663-670.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 34][Cited by in RCA: 64][Article Influence: 10.7][Reference Citation Analysis (0)]
van der Zijden CJ, Bouwman A, Mostert B, Nuyttens JJME, van der Sluis PC, Spaander MCW, Mens JWM, Homs MYV, van Doorn L, Wijnhoven BPL, Lagarde SM. Overall survival after definitive chemoradiotherapy for patients with esophageal cancer: a retrospective cohort study.Dis Esophagus. 2024;37:doae047.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 3][Reference Citation Analysis (0)]
