Clinical oncology最新文献

{"title":"Natural Language Processing to Extract Head and Neck Cancer Data From Unstructured Electronic Health Records","authors":"T. Young ,&nbsp;J. Au Yeung ,&nbsp;K. Sambasivan ,&nbsp;D. Adjogatse ,&nbsp;A. Kong ,&nbsp;I. Petkar ,&nbsp;M. Reis Ferreira ,&nbsp;M. Lei ,&nbsp;A. King ,&nbsp;J. Teo ,&nbsp;T. Guerrero Urbano","doi":"10.1016/j.clon.2025.103805","DOIUrl":"10.1016/j.clon.2025.103805","url":null,"abstract":"<div><h3>Aims</h3><div>Patient data is frequently stored as unstructured data within Electronic Health Records (EHRs), requiring manual curation. AI tools using Natural Language Processing (NLP) may rapidly curate accurate real-world unstructured EHRs to enrich datasets. We evaluated this approach for Head and Neck Cancer (HNC) patient data extraction using an open-source general-purpose healthcare NLP tool (CogStack).</div></div><div><h3>Materials and Methods</h3><div>CogStack was applied to extract relevant SNOMED-CT concepts from HNC patients' documents, generating outputs denoting the identifications of each concept for each patient. Outputs were compared to manually curated ground truth HNC datasets to calculate pre-training performance. Supervised model training was then performed using SNOMED-CT concept annotation on clinical documents, and the updated model was re-evaluated. A second training cycle was performed before the final evaluation. A thresholding approach (multiple detections needed to qualify a concept as ‘present’) was used to increase precision. The final model was evaluated on an unseen test cohort. F1 score (harmonic mean of precision and recall) was used for evaluation.</div></div><div><h3>Results</h3><div>Pre-training, the F1 score was incalculable for 19.5% of concepts due to insufficient recall. Following one training cycle, F1 score became calculable for all concepts (median 0.692). After further training, the final model demonstrated improvement in the median F1 score (0.708). Test cohort median F1 score was 0.750. Thresholding analysis developed a concept-specific best threshold approach, resulting in a median F1 score of 0.778 in the test cohort, where 50 out of 109 SNOMED-CT concepts met pre-set criteria to be considered adequately fine-tuned.</div></div><div><h3>Conclusions</h3><div>NLP can mine unstructured cancer data following limited training. Certain concepts such as histopathology terms remained poorly retrieved. Model performance is maintained when applied to a test cohort, demonstrating good generalisability. Concept-specific thresholding strategy improved performance. Fine-tuning annotations were incorporated into the NLP parent model for future performance. CogStack has been applied to extract data for 50 concepts with validated performance for our entire retrospective HNC cohort.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103805"},"PeriodicalIF":3.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143785239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prognostic Implications of HPV Cell-Free DNA Serial Testing During Follow-Up of p16 Positive Oropharyngeal Squamous Cell Carcinoma After Curative-Intent Treatment","authors":"V. Salati ,&nbsp;M. Adamowicz ,&nbsp;L. McKean ,&nbsp;D. Noble ,&nbsp;D. Srinivasan ,&nbsp;J. MacKenzie ,&nbsp;S. Linton ,&nbsp;C. Callaghan ,&nbsp;C. Robert ,&nbsp;K. Cuschieri ,&nbsp;B. Conn ,&nbsp;A. Hay ,&nbsp;T.J. Aitman ,&nbsp;I.J. Nixon","doi":"10.1016/j.clon.2025.103807","DOIUrl":"10.1016/j.clon.2025.103807","url":null,"abstract":"<div><h3>Introduction</h3><div>Plasma circulating HPV cell-free DNA has high sensitivity and specificity for the detection of HPV-mediated oropharyngeal squamous cell carcinoma. We investigated the clinical significance of serial testing after curative-intent treatments.</div></div><div><h3>Materials and Methods</h3><div>Patients with concordant p16 positive tumour or neck node biopsy and positive high-risk HPV plasma cell-free DNA were prospectively recruited. HPV cell-free DNA were obtained using digital droplet polymerase chain reaction (ddPCR) and were collected at diagnosis and at every clinical follow-up. Three months after completion of curative-intent treatments, patients were stratified according to treatment response on computed tomography. Complete responders (CR) were followed-up clinically, partial responders (PR) underwent further imaging and surgical/medical management if appropriate, patients with progressive disease (PD) received palliative treatments.</div></div><div><h3>Results</h3><div>A hundred and fourteen patients were included and 717 HPV cfDNA ddPCR samples were analysed during a median follow-up of 103 weeks (IQR, 40.2–147.8). Ninety (78.9%) patients were classified as CR, 18 (15.8%) as PR and all except one, who was rapidly diagnosed with PD, had negative HPV ddPCR at 12 weeks follow-up; 6 (5.3%) had PD and all except one had positive HPV ddPCR. Eleven had recurrent disease, 6 in the CR group (6.6%) and 5 among PR (27.7%). Ninety patients had consistently negative HPV ddPCR at all time points and one developed a recurrence (NPV 99%, 95% C.I., 93.2–99.8%). Eighteen patients developed positive HPV ddPCR and 10 developed recurrent disease (PPV 55%, 95% C.I., 38.6–71.4%). Ten patients had two consecutively positive HPV ddPCR and all had proven disease (PPV 100%, 95% C.I., 69.2–100%). Nine patients had transiently positive HPV ddPCR and none developed disease at that time.</div></div><div><h3>Conclusions</h3><div>Post-treatment HPV ddPCR reflected treatment response on imaging and serial testing had high PPV and NPV in detecting recurrent disease.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103807"},"PeriodicalIF":3.2,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143791388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Known Unknowns: Making Sense of Head and Neck Squamous Cell Carcinoma of Unknown Primary","authors":"A. Young ,&nbsp;A. Williamson ,&nbsp;J. Hardman ,&nbsp;V. Paleri ,&nbsp;B. O'Leary","doi":"10.1016/j.clon.2025.103806","DOIUrl":"10.1016/j.clon.2025.103806","url":null,"abstract":"","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103806"},"PeriodicalIF":3.2,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"OncoFlash - Research Updates in a Flash!","authors":"A. Turcas ,&nbsp;K. Thippu Jayaprakash","doi":"10.1016/j.clon.2025.103803","DOIUrl":"10.1016/j.clon.2025.103803","url":null,"abstract":"","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"40 ","pages":"Article 103803"},"PeriodicalIF":3.2,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143662966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimising HER2-positive Breast Cancer Treatment: Insights on Subcutaneous Pertuzumab-trastuzumab Transition","authors":"S. Surekha ,&nbsp;A.K. Lamiyan ,&nbsp;P. Khatri ,&nbsp;A.N. Patil","doi":"10.1016/j.clon.2025.103802","DOIUrl":"10.1016/j.clon.2025.103802","url":null,"abstract":"","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103802"},"PeriodicalIF":3.2,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence in Health Care: A Rallying Cry for Critical Clinical Research and Ethical Thinking","authors":"S.M. Bentzen","doi":"10.1016/j.clon.2025.103798","DOIUrl":"10.1016/j.clon.2025.103798","url":null,"abstract":"<div><div>Artificial intelligence (AI) will impact a large proportion of jobs in the short to medium term, especially in the developed countries. The consequences will be felt across many sectors including health care, a critical sector for implementation of AI tools because glitches in algorithms or biases in training datasets may lead to suboptimal treatment that may negatively affect the health of an individual. The stakes are obviously higher in case of potentially life-threatening diseases such as cancer and therapies with a potential for causing severe or even fatal adverse events.</div><div>Over the last two decades, much of the research on AI in health care has focussed on diagnostic radiology and digital pathology, but a solid body of research is emerging on AI tools in the radiation oncology workflow. Many of these applications are relatively uncontroversial, although there is still a lack of evidence regarding effectiveness rather than efficiency, and—the ultimate bar—evidence of clinical utility. Proponents of AI will argue that these algorithms should be implemented with robust human supervision. One challenge here is the deskilling effect associated with new technologies. We will become increasingly dependent on the AI tools over time, and we will become less capable of assessing the quality of the AI output.</div><div>Much of this research appears almost old-fashioned in view of the rapid advances in Generative artificial intelligence (GenAI). GenAI can draw from multiple types of data and produce output that is personalised and appears relevant in the given context. Especially the rapid progress in large language models (LLMs) has opened a wide field of potential applications that were out of bounds just a few years ago. One LLM, Generative Pre-trained Transformer 4 (GPT-4), has been made widely accessible to end-users as ChatGPT-4, which passed a rigorous Turing test in a recent study. In this viewpoint, I argue for the necessity of independent academic research to establish evidence-based applications of AI in medicine. Algorithmic medicine is an intervention similar to a new drug or a new medical device. We should be especially concerned about under-represented minorities and rare/atypical clinical cases that may drown in the petabyte-sized training sets. A huge educational push is needed to ensure that the end-users of AI in health care understand the strengths and weaknesses of algorithmic medicine. Finally, we need to address the ethical boundaries for where and when GenAI can replace humans in the relation between patients and healthcare providers.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103798"},"PeriodicalIF":3.2,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143758966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sexual Dysfunction in Prostate Cancer Patients According to Disease Stage and Treatment Modality","authors":"W. Kinnaird ,&nbsp;P. Schartau ,&nbsp;M. Kirby ,&nbsp;V. Jenkins ,&nbsp;S. Allen ,&nbsp;H. Payne","doi":"10.1016/j.clon.2025.103801","DOIUrl":"10.1016/j.clon.2025.103801","url":null,"abstract":"<div><h3>Aims</h3><div>To investigate physical and psychological sexual dysfunction (SD) in prostate cancer (PCa) patients, according to disease stage and treatment modality.</div></div><div><h3>Materials and methods</h3><div>Participants diagnosed with PCa completed an online survey reporting sexual side effects across 13 domains, the importance of sexual function, and their support needs. Disease stage and treatment data were collected to identify variations in experience. Results were analysed descriptively and with chi-squared significance testing.</div></div><div><h3>Results</h3><div>Six hundred fifty-four participants diagnosed with localised (66.1%), locally advanced (25.1%), and advanced (8.9%) PCa responded to the survey. Their disease management included radical prostatectomy (RP; 49.7%), radiotherapy (RT; 45.9%), and androgen deprivation therapy (ADT; 43.6%). More than 98% reported new-onset post-treatment sexual problems. The most common physical dysfunctions were erectile dysfunction (ED; 91.0%), ejaculatory disturbance (82.9%), and anatomical penile change (70.0%). The most common psychosexual dysfunctions were loss of sexual confidence (76.2%), loss of sex drive (67.1%), and loss of self-esteem (57.1%). Participants diagnosed with advanced disease were significantly more likely to report SD than participants with localised or locally advanced disease in 5 of 13 domains (p &lt; .05). Participants whose treatment included a combination of RP, RT, and ADT were most likely to report SD in 7 of 13 domains. Overall, 78.3% of participants said sexual activity was important to them, with 61.8% placing sexual problems in their top three current concerns. Furthermore, 78.3% wanted to discuss sexual problems with a healthcare professional, with most wishing to focus on ED, loss of sexual confidence, and low libido.</div></div><div><h3>Conclusion</h3><div>SD is a common, wide-ranging, and distressing side effect of treatment, and PCa survivors place a high level of importance on sexual recovery. Those with advanced disease are among the worst affected and report high levels of psychosexual problems. Holistic rehabilitation strategies addressing a broad range of side effects would benefit all, but particularly those treated with permanent ADT.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103801"},"PeriodicalIF":3.2,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Age-Specific Socioeconomic Inequalities in Treatment in Patients with Stage III Colon Cancer in England 2012–2016: A Population-Based Study with Mediation Analysis","authors":"B. Kells,&nbsp;B. Rachet,&nbsp;S. Ling","doi":"10.1016/j.clon.2025.103799","DOIUrl":"10.1016/j.clon.2025.103799","url":null,"abstract":"<div><h3>Aims</h3><div>It is unclear whether inequalities in guidelines-recommended treatment among patients with stage III colon cancer existed and differed by age in England.</div></div><div><h3>Materials and methods</h3><div>Using data from cancer registry in England between 2012 and 2016, we included all patients with stage III colon cancer and applied multivariable multinominal logistic regression, including an interaction between age and deprivation, to investigate age-specific socioeconomic inequalities in receipt of the NICE-recommend treatment – surgery combined with adjuvant chemotherapy. We also examined the mediating roles of tumour factors on the inequalities in treatment.</div></div><div><h3>Results</h3><div>Among 20,368 included patients, socioeconomic inequalities in receipt of the NICE-recommend treatment were observed at all ages but wider in patients aged between 65 and 85 years old. For a 70-year-old patient, the probability of receiving the NICE-recommend treatment was 70.8% (95% CI: 68.6, 73.1) for the least vs. 59.4% (53.7, 65.1) for the most deprived quintile. When both groups were unlikely to receive the NICE-recommended treatment (85+ years old), patients from less deprived areas had a higher probability of receiving some alternative treatments like surgery while those with the most deprived backgrounds received none. Tumour factors explained little of inequalities in receipt of surgery or adjuvant chemotherapy.</div></div><div><h3>Conclusion</h3><div>Patients from deprived areas tended to receive inferior treatment options, and tumour factors explained little of these inequalities. Guidelines need to ensure that the NICE-recommended treatment modality is available to all to reduce the survival gap.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103799"},"PeriodicalIF":3.2,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143791389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Role of Pre-ablation Stimulated Thyroglobulin and Thyroid-Stimulating Hormone Ratio for Radioactive Iodine Treatment in Adults with Papillary Thyroid Cancer","authors":"X. Zhou ,&nbsp;X. Wang ,&nbsp;W. Song ,&nbsp;X. Yue ,&nbsp;Y. Li ,&nbsp;Y. Shi","doi":"10.1016/j.clon.2025.103797","DOIUrl":"10.1016/j.clon.2025.103797","url":null,"abstract":"<div><h3>Objective</h3><div>This investigation assesses the predictive utility of the pre-ablation stimulated thyroglobulin to thyroid-stimulating hormone ratio (sTg/TSH) and examines the other factors affecting the efficacy of radioactive iodine (RAI) therapy in adult patients with papillary thyroid cancer (PTC).</div></div><div><h3>Methods</h3><div>We performed a retrospective review of clinical and pathological data from 1071 patients who received a total thyroidectomy followed by RAI therapy. The study included 576 of these patients. Participants were separated into two groups according to their reaction to RAI therapy: excellent response (ER) and non-ER (NER). The factors that contribute to NER were found using univariate and multivariate binary logistic regression analyses. The predictive importance of the sTg and sTg/TSH ratio was discovered by analyzing receiver operating characteristic (ROC) curves and setting diagnostic criteria. Decision curve analysis (DCA) was used to assess the practical implications of these findings.</div></div><div><h3>Results</h3><div>Among the 576 patients assessed, 60.07% (346 individuals) demonstrated an ER to RAI treatment. Independent predictors of a NER identified through both univariate and multivariate logistic regression analyses included multifocality (odds ratio [OR] = 2.16, 95% confidence interval [CI]: 1.28–3.67, P = 0.004), having more than ten positive lymph nodes (PLN) (OR = 3.78, 95% CI: 1.68–8.54, P = 0.001), presence of distant metastases (OR = 19.22, 95% CI: 2.09–176.93, P = 0.009), elevated stimulated thyroglobulin (sTg) levels (OR = 1.04, 95% CI: 1.00–1.07, P = 0.025), and a higher sTg/TSH ratio (OR = 2.48, 95% CI: 1.80–3.41, P &lt; 0.001). Receiver operating characteristic (ROC) curve analysis established diagnostic thresholds for predicting NER at an sTg level of 7.255 ng/ml (area under the curve [AUC] = 0.893) and an sTg/TSH ratio of 0.127 (AUC = 0.889), both demonstrating robust sensitivity and specificity. Smooth curve fitting illustrated a progressive increase in the risk of NER with rising levels of the sTg/TSH ratio. DCA confirmed the substantial clinical net benefit of these predictors in forecasting NER outcomes.</div></div><div><h3>Conclusions</h3><div>The sTg/TSH ratio is confirmed as a reliable diagnostic marker for predicting the response to primary RAI treatment in PTC. Moreover, active postoperative follow-up and surveillance are essential for patients with multifocality, PLN &gt;10, sTg &gt;7.255 ng/ml, and sTg/TSH ratio &gt;0.127.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103797"},"PeriodicalIF":3.2,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geometric and Dosimetric Evaluation of a RayStation Deep Learning Model for Auto-Segmentation of Organs at Risk in a Real-World Head and Neck Cancer Dataset","authors":"D. Sharma ,&nbsp;G. Singh ,&nbsp;N. Burela ,&nbsp;S. Gayen ,&nbsp;G. Aishwarya ,&nbsp;S. Nangia","doi":"10.1016/j.clon.2025.103796","DOIUrl":"10.1016/j.clon.2025.103796","url":null,"abstract":"<div><h3><em>Aims</em></h3><div>To assess geometric accuracy and dosimetric impact of a deep learning segmentation (DLS) model on a large, diverse dataset of head and neck cancer (HNC) patients treated with intensity-modulated proton therapy (IMPT).</div></div><div><h3><em>Materials and methods</em></h3><div>A 3D U-Net-based DLS model was applied to CT datasets of 124 HNC patients treated with IMPT at 50.4–70.0 GyRBE. Thirty organs-at-risk (OARs), delineated manually (GT-OARs) were analysed for similarity metrics with auto-segmented OARs, without (DLS-nonedited) and with (DLS-edited) manual correction, using volume, Dice similarity coefficient (DSC), and Hausdorff distance (HD). Dosimetric impact of auto-segmentation error was assessed as absolute dose difference of mean (ΔDmean) and maximum (ΔDmax).</div></div><div><h3><em>Results</em></h3><div>The cohort includes patients with postoperative (47.6%), flap reconstruction (12.1%), mouth bites (79.8%), dental implants (54.8%), and surgical implants (3.2%). DLS failed in 11 patients with significant anatomical challenges and artifact. Compared with GT-OARs, DLS-nonedited under-segmented 11/12 Gr-A (central nervous system, arteries, bone) (p &lt; 0.05) and over-segmented 13/18 Gr-B (glandular, digestive, airways) OARs. DSC scores were good (&gt;0.8), intermediate (0.6–0.8), intermediate–poor (0.5–0.6), and poor (&lt;0.5) in 12, 6, 4, and 8 OARs. HD were good (&lt;4mm), intermediate (4–6mm), poor (6–8mm), and very poor (&gt;8mm) in 5, 7, 4, and 14 OARs. Compared with manually corrected, DLS-edited OARs, all DLS-nonedited OARs demonstrated excellent similarity with DSC&gt;0.8 and HD&lt;4mm. On average, auto-segmentation took 2.51 minutes, while correction took 6.24 minutes. The mean values of ΔDmean and ΔDmax were within ±300 and ±3 cGyRBE, except for oesophagus and larynx, where the mean ΔDmean increases up to 837.14 cGyRBE.</div></div><div><h3><em>Conclusion</em></h3><div>Patient posture, nonbiological materials, and anatomical deformities influence DLS accuracy. The model’s overall performance is adequate and efficient with skilled manual editing needed for few OARs.</div></div>","PeriodicalId":10403,"journal":{"name":"Clinical oncology","volume":"41 ","pages":"Article 103796"},"PeriodicalIF":3.2,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}