Clinical Imaging最新文献_第5页

Multimodal approach to optimize biopsy decision-making for PI-RADS 3 lesions on multiparametric MRI 多模态方法优化多参数磁共振成像上 PI-RADS 3 病变的活检决策。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-19 DOI: 10.1016/j.clinimag.2024.110363

Omer Tarik Esengur , Enis C. Yilmaz , Kutsev B. Ozyoruk , Alex Chen , Nathan S. Lay , David G. Gelikman , Maria J. Merino , Sandeep Gurram , Bradford J. Wood , Peter L. Choyke , Stephanie A. Harmon , Peter A. Pinto , Baris Turkbey

Purpose

To develop and evaluate a multimodal approach including clinical parameters and biparametric MRI-based artificial intelligence (AI) model for determining the necessity of prostate biopsy in patients with PI-RADS 3 lesions.

Methods

This retrospective study included a prospectively recruited patient cohort with PI-RADS 3 lesions who underwent prostate MRI and MRI/US fusion-guided biopsy between April 2019 and February 2024 in a single institution. The study examined demographic data, PSA and PSA density (PSAD) levels, prostate volumes, prospective PI-RADS v2.1-compliant interpretations of a genitourinary radiologist, lesion characteristics, history of prior biopsies, and AI evaluations, focusing mainly on the detection of clinically significant prostate cancer (csPCa) (International Society of Urological Pathology grade group ≥2) on MRI/US fusion-guided biopsy. The AI model lesion segmentations were compared to manual segmentations and biopsy results. The statistical methods employed included Fisher's exact test and logistic regression.

Results

The cohort was comprised of 248 patients with 312 PI-RADS 3 lesions in total (n = 268 non-csPCa, n = 44 csPCa). The AI model's negative predictive value (NPV) was 89.2 % for csPCa in all lesions. In patient-level analysis, the NPV was 91.2 % for patients with a highest PI-RADS score of 3. PSAD was a significant predictor of csPCa (odds ratio = 5.8, p = 0.038). Combining AI and PSAD, where AI correctly mapped a lesion or PSAD ≥0.15 ng/mL², achieved higher sensitivity (77.8 %) while maintaining a high NPV (93.1 %).

Conclusion

Combining AI and PSAD has the potential to enhance biopsy decision-making for PI-RADS 3 lesions by minimizing missed csPCa occurrences and reducing unnecessary biopsies.

目的：开发和评估一种多模态方法，包括临床参数和基于双参数 MRI 的人工智能（AI）模型，用于确定 PI-RADS 3 病变患者进行前列腺活检的必要性：这项回顾性研究纳入了前瞻性招募的PI-RADS 3病变患者队列，这些患者在2019年4月至2024年2月期间在一家机构接受了前列腺MRI和MRI/US融合引导下的活检。该研究检查了人口统计学数据、PSA和PSA密度（PSAD）水平、前列腺体积、泌尿生殖系统放射科医生符合PI-RADS v2.1标准的前瞻性解释、病变特征、既往活检史和AI评估，主要侧重于在MRI/US融合引导活检中发现有临床意义的前列腺癌（csPCa）（国际泌尿病理学会分级组≥2）。人工智能模型病灶分割与人工分割和活检结果进行了比较。采用的统计方法包括费雪精确检验和逻辑回归：结果：研究组由 248 名患者组成，共有 312 个 PI-RADS 3 病灶（n = 268 个非 csPCa，n = 44 个 csPCa）。在所有病变中，AI 模型对 csPCa 的阴性预测值 (NPV) 为 89.2%。在患者层面的分析中，PI-RADS 最高评分为 3 分的患者的阴性预测值为 91.2%。PSAD 是 csPCa 的重要预测因子（几率比 = 5.8，p = 0.038）。结合 AI 和 PSAD，当 AI 正确映射出病变或 PSAD ≥0.15 ng/mL2 时，可获得更高的灵敏度（77.8%），同时保持较高的 NPV（93.1%）：结论：将 AI 和 PSAD 结合使用可最大限度地减少漏诊 csPCa 并减少不必要的活检，从而提高 PI-RADS 3 病变的活检决策水平。

{"title":"Multimodal approach to optimize biopsy decision-making for PI-RADS 3 lesions on multiparametric MRI","authors":"Omer Tarik Esengur , Enis C. Yilmaz , Kutsev B. Ozyoruk , Alex Chen , Nathan S. Lay , David G. Gelikman , Maria J. Merino , Sandeep Gurram , Bradford J. Wood , Peter L. Choyke , Stephanie A. Harmon , Peter A. Pinto , Baris Turkbey","doi":"10.1016/j.clinimag.2024.110363","DOIUrl":"10.1016/j.clinimag.2024.110363","url":null,"abstract":"<div><h3>Purpose</h3><div>To develop and evaluate a multimodal approach including clinical parameters and biparametric MRI-based artificial intelligence (AI) model for determining the necessity of prostate biopsy in patients with PI-RADS 3 lesions.</div></div><div><h3>Methods</h3><div>This retrospective study included a prospectively recruited patient cohort with PI-RADS 3 lesions who underwent prostate MRI and MRI/US fusion-guided biopsy between April 2019 and February 2024 in a single institution. The study examined demographic data, PSA and PSA density (PSAD) levels, prostate volumes, prospective PI-RADS v2.1-compliant interpretations of a genitourinary radiologist, lesion characteristics, history of prior biopsies, and AI evaluations, focusing mainly on the detection of clinically significant prostate cancer (csPCa) (International Society of Urological Pathology grade group ≥2) on MRI/US fusion-guided biopsy. The AI model lesion segmentations were compared to manual segmentations and biopsy results. The statistical methods employed included Fisher's exact test and logistic regression.</div></div><div><h3>Results</h3><div>The cohort was comprised of 248 patients with 312 PI-RADS 3 lesions in total (<em>n</em> = 268 non-csPCa, <em>n</em> = 44 csPCa). The AI model's negative predictive value (NPV) was 89.2 % for csPCa in all lesions. In patient-level analysis, the NPV was 91.2 % for patients with a highest PI-RADS score of 3. PSAD was a significant predictor of csPCa (odds ratio = 5.8, <em>p</em> = 0.038). Combining AI and PSAD, where AI correctly mapped a lesion or PSAD ≥0.15 ng/mL<sup>2</sup>, achieved higher sensitivity (77.8 %) while maintaining a high NPV (93.1 %).</div></div><div><h3>Conclusion</h3><div>Combining AI and PSAD has the potential to enhance biopsy decision-making for PI-RADS 3 lesions by minimizing missed csPCa occurrences and reducing unnecessary biopsies.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"117 ","pages":"Article 110363"},"PeriodicalIF":1.8,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142695999","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Short-term treatment response assessment in non-surgical treatment of advanced non-small cell lung cancer based on radiomics of dual-energy CT 基于双能 CT 放射组学的晚期非小细胞肺癌非手术治疗的短期治疗反应评估。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-19 DOI: 10.1016/j.clinimag.2024.110362

Xiuting Wu , Yumin Lu , Danmei Huang , Zefeng Li , Chunchen Wei , Kai Li

Purpose

To build and evaluate a pre-treatment dual-energy CT(DECT)-based clinical-radiomics nomogram for individualized prediction of short-term treatment response to non-surgical treatment in advanced non-small cell lung cancer (NSCLC).

Methods

Pre-treatment DECT images were retrospectively collected from 98 pathologically confirmed NSCLC with clinical stage III or IV. Short-term treatment response was determined with follow-up CT of 4–6 courses of treatment. Quantitative radiomics metrics of the lesion were extracted from dual-energy mixed images at venous phase. Least absolute shrinkage and selection operator and correlation analysis were used to select the most relevant radiomics features. Radiomics model, clinical model and clinical-radiomics model were established by multivariate logistic regression. The model with the best prediction performance was visualized as a nomogram, and the consistency between the probability of the actual occurrence of the outcome and the probability predicted by the model was measured by calibration curves.

Results

Clinical stage, difference in electron density in arteriovenous phase, difference in slope of energy spectrum in arteriovenous phase, and slope of energy spectrum in venous phase of the tumor were significant clinical predictors of therapy response (P < 0.05). The clinical-radiomics model showed a higher predictive capability (AUC: 0.87 and 0.85 in training and validation sets, respectively) than the radiomics models and the clinical model. The clinical-radiomics nomogram integrating the DECT radiomics signature with clinical stage and spectrum parameters showed good calibration and discrimination.

Conclusion

The clinical-radiomics nomogram based on pre-treatment DECT showed good performance in predicting clinical response to non-surgical therapy in NSCLC.

目的：建立并评估基于治疗前双能 CT（DECT）的临床放射组学提名图，用于个体化预测晚期非小细胞肺癌（NSCLC）非手术治疗的短期治疗反应：方法：回顾性收集了98例经病理证实为临床III期或IV期的NSCLC患者的治疗前DECT图像。通过 4-6 个疗程的随访 CT 确定短期治疗反应。从静脉期双能混合图像中提取病变的定量放射组学指标。使用最小绝对收缩和选择算子以及相关性分析来选择最相关的放射组学特征。通过多元逻辑回归建立了放射组学模型、临床模型和临床-放射组学模型。将预测效果最好的模型以提名图的形式直观显示，并通过校准曲线测量结果实际发生概率与模型预测概率之间的一致性：结果：临床分期、动静脉期电子密度差异、动静脉期能谱斜率差异和静脉期能谱斜率是治疗反应的显著临床预测因子（P 结论：临床放射组学提名图是治疗反应的显著临床预测因子：基于治疗前 DECT 的临床放射组学提名图在预测 NSCLC 非手术治疗的临床反应方面表现良好。

{"title":"Short-term treatment response assessment in non-surgical treatment of advanced non-small cell lung cancer based on radiomics of dual-energy CT","authors":"Xiuting Wu , Yumin Lu , Danmei Huang , Zefeng Li , Chunchen Wei , Kai Li","doi":"10.1016/j.clinimag.2024.110362","DOIUrl":"10.1016/j.clinimag.2024.110362","url":null,"abstract":"<div><h3>Purpose</h3><div>To build and evaluate a pre-treatment dual-energy CT(DECT)-based clinical-radiomics nomogram for individualized prediction of short-term treatment response to non-surgical treatment in advanced non-small cell lung cancer (NSCLC).</div></div><div><h3>Methods</h3><div>Pre-treatment DECT images were retrospectively collected from 98 pathologically confirmed NSCLC with clinical stage III or IV. Short-term treatment response was determined with follow-up CT of 4–6 courses of treatment. Quantitative radiomics metrics of the lesion were extracted from dual-energy mixed images at venous phase. Least absolute shrinkage and selection operator and correlation analysis were used to select the most relevant radiomics features. Radiomics model, clinical model and clinical-radiomics model were established by multivariate logistic regression. The model with the best prediction performance was visualized as a nomogram, and the consistency between the probability of the actual occurrence of the outcome and the probability predicted by the model was measured by calibration curves.</div></div><div><h3>Results</h3><div>Clinical stage, difference in electron density in arteriovenous phase, difference in slope of energy spectrum in arteriovenous phase, and slope of energy spectrum in venous phase of the tumor were significant clinical predictors of therapy response (<em>P</em> < 0.05). The clinical-radiomics model showed a higher predictive capability (AUC: 0.87 and 0.85 in training and validation sets, respectively) than the radiomics models and the clinical model. The clinical-radiomics nomogram integrating the DECT radiomics signature with clinical stage and spectrum parameters showed good calibration and discrimination.</div></div><div><h3>Conclusion</h3><div>The clinical-radiomics nomogram based on pre-treatment DECT showed good performance in predicting clinical response to non-surgical therapy in NSCLC.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"117 ","pages":"Article 110362"},"PeriodicalIF":1.8,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142693791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reply to: “Comment: Radiologists’ Perspectives on AI and Opportunistic CT Screening (OS)” 答复"评论：放射科医生对人工智能和机会性 CT 筛查的看法 (OS)"。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-17 DOI: 10.1016/j.clinimag.2024.110359

Adam E.M. Eltorai , Katherine P. Andriole

引用次数: 0

Low breast density is associated with epicardial adipose tissue volume and coronary artery disease 低乳房密度与心外膜脂肪组织体积和冠状动脉疾病有关。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-17 DOI: 10.1016/j.clinimag.2024.110357

Emma Aldous , Vinay Goel , Chee Yeong , Nushrat Sultana , Rachael Hii , Huong Tu , Anthony Salib , Edwin Xu , Sarang Paleri , Sheran Vasanthakumar , Rhea Nandurkar , Andrew Lin , Nitesh Nerlekar

Purpose

Epicardial adipose tissue volume (EATv), is well correlated with coronary artery disease (CAD), however not reported clinically. Breast density, measured on mammography, has shown promise as a reflector of cardiometabolic risk, with less dense breasts indicating greater proportion of adipose tissue. We aimed to evaluate the association between breast density, EATv and CAD.

Method

Retrospective, cross-sectional study including 153 women who had both clinically indicated coronary computed tomography angiogram (CCTA) and mammography. EATv was quantified using semi-automated software. Breast density was visually assessed by standard 4-level BI-RADS grading (low: BI-RADS A–B, high: BI-RADS CD). CAD was categorised as presence/absence of coronary artery plaque and severity was quantified using CAD-RADS score.

Results

Among 153 patients (mean age 62 ± 10), 103 (67.3 %) had low breast density (high breast adiposity). Low breast density patients were older, had greater rates of hypertension, higher mean BMI (p < 0.001) and EATv (106.6 ± 43.0 ml vs 81.0 ± 31.6 ml, p < 0.001). EATv was predictive of low breast density (OR: 1.02[1.01–1.03], p = 0.006), independent of age and hypertension. Low breast density was strongly associated with presence of CAD (prevalence 75 % vs 48 %, OR: 3.21[1.58–6.53], p = 0.001) independent of EATv, and modifiable (OR: 2.69[1.24–5.92], p = 0.012) and non-modifiable (OR: 2.42[1.04–5.85], p = 0.047) cardiovascular risk factors. Low breast density made up a higher proportion of mild (76.5 %), moderate (73.9 %) and severe (80.0 %) CAD.

Conclusions

Low breast density is associated with higher EATv and independently associated with CAD presence beyond EATv and other cardiovascular risk factors. Mammographic breast density may therefore have value as an early risk identification tool for CAD in women.

目的：心外膜脂肪组织体积（EATv）与冠状动脉疾病（CAD）有很好的相关性，但没有临床报道。通过乳房 X 射线照相术测量的乳房密度有望反映心脏代谢风险，密度较低的乳房表明脂肪组织的比例较高。我们的目的是评估乳房密度、EATv 和 CAD 之间的关联：方法：回顾性横断面研究，包括 153 名接受过临床指征冠状动脉计算机断层扫描（CCTA）和乳房 X 射线照相术的女性。使用半自动软件对 EATv 进行量化。乳腺密度通过标准的四级 BI-RADS 分级（低：BI-RADS A-B，高：BI-RADS CD）进行目测评估。冠状动脉粥样硬化分为有/无冠状动脉斑块，严重程度用 CAD-RADS 评分量化：在 153 名患者（平均年龄 62 ± 10 岁）中，103 人（67.3%）乳房密度低（乳房脂肪含量高）。低乳腺密度患者年龄更大，高血压发病率更高，平均体重指数（BMI）更高（P 结论：低乳腺密度与更高的 EEG 相关：低乳房密度与较高的 EATv 有关，并且与 CAD 的存在有独立的关联，超出了 EATv 和其他心血管风险因素。因此，乳腺X线照相术中的乳腺密度可作为早期识别女性患 CAD 风险的工具。

{"title":"Low breast density is associated with epicardial adipose tissue volume and coronary artery disease","authors":"Emma Aldous , Vinay Goel , Chee Yeong , Nushrat Sultana , Rachael Hii , Huong Tu , Anthony Salib , Edwin Xu , Sarang Paleri , Sheran Vasanthakumar , Rhea Nandurkar , Andrew Lin , Nitesh Nerlekar","doi":"10.1016/j.clinimag.2024.110357","DOIUrl":"10.1016/j.clinimag.2024.110357","url":null,"abstract":"<div><h3>Purpose</h3><div>Epicardial adipose tissue volume (EATv), is well correlated with coronary artery disease (CAD), however not reported clinically. Breast density, measured on mammography, has shown promise as a reflector of cardiometabolic risk, with less dense breasts indicating greater proportion of adipose tissue. We aimed to evaluate the association between breast density, EATv and CAD.</div></div><div><h3>Method</h3><div>Retrospective, cross-sectional study including 153 women who had both clinically indicated coronary computed tomography angiogram (CCTA) and mammography. EATv was quantified using semi-automated software. Breast density was visually assessed by standard 4-level BI-RADS grading (low: BI-RADS A–B, high: BI-RADS C<img>D). CAD was categorised as presence/absence of coronary artery plaque and severity was quantified using CAD-RADS score.</div></div><div><h3>Results</h3><div>Among 153 patients (mean age 62 ± 10), 103 (67.3 %) had low breast density (high breast adiposity). Low breast density patients were older, had greater rates of hypertension, higher mean BMI (<em>p</em> <em><</em> 0.001) and EATv (106.6 ± 43.0 ml vs 81.0 ± 31.6 ml, <em>p</em> <em><</em> 0.001). EATv was predictive of low breast density (OR: 1.02[1.01–1.03], <em>p</em> <em>=</em> 0.006), independent of age and hypertension. Low breast density was strongly associated with presence of CAD (prevalence 75 % vs 48 %, OR: 3.21[1.58–6.53], <em>p</em> <em>=</em> 0.001) independent of EATv, and modifiable (OR: 2.69[1.24–5.92], <em>p</em> <em>=</em> 0.012) and non-modifiable (OR: 2.42[1.04–5.85], <em>p</em> <em>=</em> 0.047) cardiovascular risk factors. Low breast density made up a higher proportion of mild (76.5 %), moderate (73.9 %) and severe (80.0 %) CAD.</div></div><div><h3>Conclusions</h3><div>Low breast density is associated with higher EATv and independently associated with CAD presence beyond EATv and other cardiovascular risk factors. Mammographic breast density may therefore have value as an early risk identification tool for CAD in women.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"117 ","pages":"Article 110357"},"PeriodicalIF":1.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142683215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Comment: Radiologists' perspectives on AI and opportunistic CT screening (OS) 评论：放射科医生对人工智能和机会性 CT 筛查的看法 (OS)。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-17 DOI: 10.1016/j.clinimag.2024.110360

Md. Tauseef Qamar, Juhi Yasmeen

引用次数: 0

Debunking a mythology: Atelectasis is not a cause of postoperative fever 揭穿神话：气胸不是术后发烧的原因。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-17 DOI: 10.1016/j.clinimag.2024.110358

Hadassah Stein , John Denning , Huma Ahmed , Michael A. Bruno , Marc Gosselin , Jinel Scott , Stephen Waite

Most physicians appreciate that practicing medicine is a commitment to continuous learning. However, “learning” can be mistakenly understood as simply the acquisition of facts and new knowledge. But learning also necessitates the constant re-examination and challenging of one's existing body of knowledge, as misinformation persists when one's beliefs are not challenged or questioned in the light of new information. One example is the pervasive belief that postoperative atelectasis causes fever despite ample evidence to the contrary. Herein we examine the imaging characteristics of atelectasis, and the means of differentiation of atelectasis from consolidation. We also explore the history of this persistent myth and review the existing literature on the actual causes of postoperative fever.

大多数医生都明白，行医就是不断学习的承诺。然而，"学习 "可能被错误地理解为仅仅是获取事实和新知识。但是，学习也需要不断重新审视和挑战自己现有的知识体系，因为如果不根据新信息对自己的信念提出挑战或质疑，错误信息就会持续存在。其中一个例子是，尽管有大量证据表明术后气胸会导致发热，但人们普遍认为气胸会导致发热。在此，我们研究了肺不张的影像学特征，以及区分肺不张和肺凝固的方法。我们还探讨了这一顽固说法的历史，并回顾了有关术后发热实际原因的现有文献。

引用次数: 0

Cultivating diagnostic clarity: The importance of reporting artificial intelligence confidence levels in radiologic diagnoses 培养诊断清晰度：在放射诊断中报告人工智能置信度的重要性。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-13 DOI: 10.1016/j.clinimag.2024.110356

Mobina Fathi , Kimia Vakili , Ramtin Hajibeygi , Ashkan Bahrami , Shima Behzad , Armin Tafazolimoghadam , Hadiseh Aghabozorgi , Reza Eshraghi , Vivek Bhatt , Ali Gholamrezanezhad

Accurate image interpretation is essential in the field of radiology to the healthcare team in order to provide optimal patient care. This article discusses the use of artificial intelligence (AI) confidence levels to enhance the accuracy and dependability of its radiological diagnoses. The current advances in AI technologies have changed how radiologists and clinicians make the diagnoses of pathological conditions such as aneurysms, hemorrhages, pneumothorax, pneumoperitoneum, and particularly fractures.

To enhance the utility of these AI models, radiologists need a more comprehensive understanding of the model's levels of confidence and certainty behind the results they produce. This allows radiologists to make more informed decisions that have the potential to drastically change a patient's clinical management. Several AI models, especially those utilizing deep learning models (DL) with convolutional neural networks (CNNs), have demonstrated significant potential in identifying subtle findings in medical imaging that are often missed by radiologists.

It is necessary to create standardized levels of confidence metrics in order for AI systems to be relevant and reliable in the clinical setting. Incorporating AI into clinical practice does have certain obstacles like the need for clinical validation, concerns regarding the interpretability of AI system results, and addressing confusion and misunderstandings within the medical community. This study emphasizes the importance of AI systems to clearly convey their level of confidence in radiological diagnosis. This paper highlights the importance of conducting research to establish AI confidence level metrics that are limited to a specific anatomical region or lesion type.

Key point of the view

Accurate fracture diagnosis relies on radiologic certainty, where Artificial intelligence (AI), especially convolutional neural networks (CNNs) and deep learning (DL), shows promise in enhancing X-ray interpretation amidst a shortage of radiologists. Overcoming integration challenges through improved AI interpretability and education is crucial for widespread acceptance and better patient outcomes.

在放射学领域，准确的图像解读对医疗团队提供最佳的病人护理至关重要。本文讨论了如何利用人工智能（AI）置信度来提高放射诊断的准确性和可靠性。目前，人工智能技术的进步已经改变了放射科医生和临床医生对动脉瘤、出血、气胸、腹腔积气，尤其是骨折等病理情况的诊断方式。为了提高这些人工智能模型的实用性，放射科医生需要更全面地了解模型所产生结果背后的置信度和确定性。这能让放射科医生做出更明智的决定，而这些决定有可能极大地改变患者的临床管理。一些人工智能模型，特别是那些利用卷积神经网络（CNN）的深度学习模型（DL），在识别医学影像中经常被放射科医生遗漏的细微发现方面表现出了巨大的潜力。有必要建立标准化的置信度指标，以使人工智能系统在临床环境中具有相关性和可靠性。将人工智能融入临床实践确实存在一些障碍，如需要临床验证、对人工智能系统结果可解释性的担忧，以及解决医学界的困惑和误解。本研究强调了人工智能系统明确表达其在放射诊断中的可信度的重要性。本文强调了开展研究以建立仅限于特定解剖区域或病变类型的人工智能置信度指标的重要性。观点要点：准确的骨折诊断依赖于放射学的确定性，而人工智能（AI），尤其是卷积神经网络（CNN）和深度学习（DL），在放射科医生短缺的情况下有望提高 X 射线判读能力。通过提高人工智能的可解释性和教育来克服整合方面的挑战，对于获得广泛认可和改善患者治疗效果至关重要。

{"title":"Cultivating diagnostic clarity: The importance of reporting artificial intelligence confidence levels in radiologic diagnoses","authors":"Mobina Fathi , Kimia Vakili , Ramtin Hajibeygi , Ashkan Bahrami , Shima Behzad , Armin Tafazolimoghadam , Hadiseh Aghabozorgi , Reza Eshraghi , Vivek Bhatt , Ali Gholamrezanezhad","doi":"10.1016/j.clinimag.2024.110356","DOIUrl":"10.1016/j.clinimag.2024.110356","url":null,"abstract":"<div><div>Accurate image interpretation is essential in the field of radiology to the healthcare team in order to provide optimal patient care. This article discusses the use of artificial intelligence (AI) confidence levels to enhance the accuracy and dependability of its radiological diagnoses. The current advances in AI technologies have changed how radiologists and clinicians make the diagnoses of pathological conditions such as aneurysms, hemorrhages, pneumothorax, pneumoperitoneum, and particularly fractures.</div><div>To enhance the utility of these AI models, radiologists need a more comprehensive understanding of the model's levels of confidence and certainty behind the results they produce. This allows radiologists to make more informed decisions that have the potential to drastically change a patient's clinical management. Several AI models, especially those utilizing deep learning models (DL) with convolutional neural networks (CNNs), have demonstrated significant potential in identifying subtle findings in medical imaging that are often missed by radiologists.</div><div>It is necessary to create standardized levels of confidence metrics in order for AI systems to be relevant and reliable in the clinical setting. Incorporating AI into clinical practice does have certain obstacles like the need for clinical validation, concerns regarding the interpretability of AI system results, and addressing confusion and misunderstandings within the medical community. This study emphasizes the importance of AI systems to clearly convey their level of confidence in radiological diagnosis. This paper highlights the importance of conducting research to establish AI confidence level metrics that are limited to a specific anatomical region or lesion type.</div></div><div><h3>Key point of the view</h3><div>Accurate fracture diagnosis relies on radiologic certainty, where Artificial intelligence (AI), especially convolutional neural networks (CNNs) and deep learning (DL), shows promise in enhancing X-ray interpretation amidst a shortage of radiologists. Overcoming integration challenges through improved AI interpretability and education is crucial for widespread acceptance and better patient outcomes.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"117 ","pages":"Article 110356"},"PeriodicalIF":1.8,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142683186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Diagnostic performance of an artificial intelligence model for the detection of pneumothorax at chest X-ray 人工智能模型在胸部 X 光检测气胸方面的诊断性能。

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-12 DOI: 10.1016/j.clinimag.2024.110355

Caterina Beatrice Monti , Lorenzo Maria Giuseppe Bianchi , Francesco Rizzetto , Luca Alessandro Carbonaro , Angelo Vanzulli

Purpose

Pneumothorax (PTX) is a common clinical urgency, its diagnosis is usually performed on chest radiography (CXR), and it presents a setting where artificial intelligence (AI) methods could find terrain in aiding radiologists in facing increasing workloads. Hence, the purpose of our study was to test an AI system for the detection of PTX on CXR examinations, to review its diagnostic performance in such setting alongside that of reading radiologists.

Method

We retrospectively ran an AI system on CXR examinations of patients who were imaged for the suspicion of PTX, and who also underwent computed tomography (CT) within the same day, the latter being used as reference standard. The performance of the proposed AI system was compared to that of reading radiologists, obtained from CXR reports.

Results

Overall, the AI system achieved an accuracy of 74 % (95%CI 68–79 %), with a sensitivity of 66 % (95%CI 59–73 %) and a specificity of 93 % (95%CI 85–97 %). Human readers displayed a comparable accuracy (77 %, 95%CI 71–82 %, p = 0.355), with higher sensitivity (73 %, 95%CI 66–79 %, p = 0.040), albeit lower specificity (85 %, 95%CI 75–91 %, p = 0.034). The performance of AI was influenced by patient positioning at CXR (p = 0.040).

Conclusions

The proposed tool could represent an aid to radiologists in detecting PTX, improving specificity. Further improvement with training on more challenging cases may pave the way for its use as a screening or standalone tool.

目的：气胸（PTX）是一种常见的临床急症，其诊断通常是通过胸部X光检查（CXR）进行的，人工智能（AI）方法可以在这种情况下帮助放射科医生应对日益增加的工作量。因此，我们研究的目的是测试一个在 CXR 检查中检测 PTX 的人工智能系统，以审查其在这种情况下与放射科医生的诊断性能：方法：我们对因怀疑患有 PTX 而接受造影检查的患者的 CXR 检查进行了回顾性人工智能系统测试，这些患者在同一天还接受了计算机断层扫描（CT）检查，后者被用作参考标准。将所建议的人工智能系统的性能与阅片放射科医生从 CXR 报告中获得的性能进行了比较：总体而言，人工智能系统的准确率为 74%（95%CI 68-79%），灵敏度为 66%（95%CI 59-73%），特异性为 93%（95%CI 85-97%）。人类读者的准确率相当（77 %，95%CI 71-82 %，p = 0.355），灵敏度较高（73 %，95%CI 66-79 %，p = 0.040），但特异性较低（85 %，95%CI 75-91 %，p = 0.034）。人工智能的性能受患者进行 CXR 检查时的体位影响（p = 0.040）：结论：所提出的工具可帮助放射科医生检测 PTX，提高特异性。通过对更具挑战性的病例进行培训来进一步改进该工具，可为其作为筛查工具或独立工具使用铺平道路。

{"title":"Diagnostic performance of an artificial intelligence model for the detection of pneumothorax at chest X-ray","authors":"Caterina Beatrice Monti , Lorenzo Maria Giuseppe Bianchi , Francesco Rizzetto , Luca Alessandro Carbonaro , Angelo Vanzulli","doi":"10.1016/j.clinimag.2024.110355","DOIUrl":"10.1016/j.clinimag.2024.110355","url":null,"abstract":"<div><h3>Purpose</h3><div>Pneumothorax (PTX) is a common clinical urgency, its diagnosis is usually performed on chest radiography (CXR), and it presents a setting where artificial intelligence (AI) methods could find terrain in aiding radiologists in facing increasing workloads. Hence, the purpose of our study was to test an AI system for the detection of PTX on CXR examinations, to review its diagnostic performance in such setting alongside that of reading radiologists.</div></div><div><h3>Method</h3><div>We retrospectively ran an AI system on CXR examinations of patients who were imaged for the suspicion of PTX, and who also underwent computed tomography (CT) within the same day, the latter being used as reference standard. The performance of the proposed AI system was compared to that of reading radiologists, obtained from CXR reports.</div></div><div><h3>Results</h3><div>Overall, the AI system achieved an accuracy of 74 % (95%CI 68–79 %), with a sensitivity of 66 % (95%CI 59–73 %) and a specificity of 93 % (95%CI 85–97 %). Human readers displayed a comparable accuracy (77 %, 95%CI 71–82 %, <em>p</em> = 0.355), with higher sensitivity (73 %, 95%CI 66–79 %, <em>p</em> = 0.040), albeit lower specificity (85 %, 95%CI 75–91 %, <em>p</em> = 0.034). The performance of AI was influenced by patient positioning at CXR (<em>p</em> = 0.040).</div></div><div><h3>Conclusions</h3><div>The proposed tool could represent an aid to radiologists in detecting PTX, improving specificity. Further improvement with training on more challenging cases may pave the way for its use as a screening or standalone tool.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"117 ","pages":"Article 110355"},"PeriodicalIF":1.8,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142677538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Minimizing prostate diffusion weighted MRI examination time through deep learning reconstruction 通过深度学习重建最大限度缩短前列腺弥散加权磁共振成像检查时间

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-05 DOI: 10.1016/j.clinimag.2024.110341

Rory L. Cochran , Eugene Milshteyn , Soumyadeep Ghosh , Nabih Nakrour , Nathaniel D. Mercaldo , Arnaud Guidon , Mukesh G. Harisinghani

Purpose

To study the diagnostic image quality of high b-value diffusion weighted images (DWI) derived from standard and variably reduced datasets reconstructed with a commercially available deep learning reconstruction (DLR) algorithm.

Materials and methods

This was a retrospective study of 52 patients undergoing conventional prostate MRI with raw image data reconstructed using both conventional 2D Cartesian and DLR algorithms. Simulated shortened DWI acquisition times were performed by reconstructing images using DLR datasets harboring a reduced number of excitations (NEX). Two radiologists independently evaluated the image quality using a 4-point Likert scale. Signal-to-noise ratio (SNR) analysis was performed for the entire cohort and a subset of patients identified as having clinically significant prostate cancer identified at MRI, and later confirmed by pathology.

Results

Radiologists perceived less image noise for both restricted and large field of view (FOV) standard NEX dataset DLR diffusion images compared to conventionally reconstructed images with good interreader agreement. Diagnostic image quality was maintained for all DLR images using variably reduced NEX compared to conventionally reconstructed images employing the standard NEX. Improved signal to noise was observed for the restricted FOV DLR images compared to conventional reconstruction using standard NEX. DLR diffusion images derived from reduced NEX datasets translated to time reductions of up to 68 % and 39 % for the restricted and large FOV series acquisitions, respectively.

Conclusion

DLR of diffusion weighted images can reduce image noise at standard NEX and potentially reduce prostate MRI exam time when utilizing reduced NEX datasets without sacrificing diagnostic image quality.

目的：研究使用市售深度学习重建（DLR）算法重建的标准数据集和可变缩小数据集生成的高b值弥散加权图像（DWI）的诊断图像质量：这是一项回顾性研究，研究对象是 52 名接受传统前列腺 MRI 检查的患者，原始图像数据采用传统的二维笛卡尔算法和 DLR 算法重建。通过使用包含较少激发数（NEX）的 DLR 数据集重建图像，模拟缩短了 DWI 采集时间。两名放射科医生使用 4 点李克特量表独立评估图像质量。信噪比（SNR）分析针对的是整个组群，以及在核磁共振成像中被确定为患有临床意义重大的前列腺癌并随后经病理证实的患者子集：与传统的重建图像相比，放射科医生认为限制视野和大视野（FOV）标准 NEX 数据集 DLR 扩散图像的噪声较小，读片者之间的一致性良好。与使用标准 NEX 的传统重建图像相比，使用变异缩小 NEX 的所有 DLR 图像都能保持诊断图像质量。与使用标准 NEX 的传统重建相比，受限 FOV DLR 图像的信噪比有所改善。从缩小的 NEX 数据集中得到的 DLR 扩散图像可使受限 FOV 和大 FOV 系列采集的时间分别缩短 68% 和 39%：弥散加权图像的 DLR 可以减少标准 NEX 的图像噪声，在不影响诊断图像质量的情况下，利用缩小 NEX 数据集有可能缩短前列腺 MRI 检查时间。

{"title":"Minimizing prostate diffusion weighted MRI examination time through deep learning reconstruction","authors":"Rory L. Cochran , Eugene Milshteyn , Soumyadeep Ghosh , Nabih Nakrour , Nathaniel D. Mercaldo , Arnaud Guidon , Mukesh G. Harisinghani","doi":"10.1016/j.clinimag.2024.110341","DOIUrl":"10.1016/j.clinimag.2024.110341","url":null,"abstract":"<div><h3>Purpose</h3><div>To study the diagnostic image quality of high b-value diffusion weighted images (DWI) derived from standard and variably reduced datasets reconstructed with a commercially available deep learning reconstruction (DLR) algorithm.</div></div><div><h3>Materials and methods</h3><div>This was a retrospective study of 52 patients undergoing conventional prostate MRI with raw image data reconstructed using both conventional 2D Cartesian and DLR algorithms. Simulated shortened DWI acquisition times were performed by reconstructing images using DLR datasets harboring a reduced number of excitations (NEX). Two radiologists independently evaluated the image quality using a 4-point Likert scale. Signal-to-noise ratio (SNR) analysis was performed for the entire cohort and a subset of patients identified as having clinically significant prostate cancer identified at MRI, and later confirmed by pathology.</div></div><div><h3>Results</h3><div>Radiologists perceived less image noise for both restricted and large field of view (FOV) standard NEX dataset DLR diffusion images compared to conventionally reconstructed images with good interreader agreement. Diagnostic image quality was maintained for all DLR images using variably reduced NEX compared to conventionally reconstructed images employing the standard NEX. Improved signal to noise was observed for the restricted FOV DLR images compared to conventional reconstruction using standard NEX. DLR diffusion images derived from reduced NEX datasets translated to time reductions of up to 68 % and 39 % for the restricted and large FOV series acquisitions, respectively.</div></div><div><h3>Conclusion</h3><div>DLR of diffusion weighted images can reduce image noise at standard NEX and potentially reduce prostate MRI exam time when utilizing reduced NEX datasets without sacrificing diagnostic image quality.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"117 ","pages":"Article 110341"},"PeriodicalIF":1.8,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142631855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Beyond the surface: A comprehensive radiological review of primary retroperitoneal neoplasms 超越表面：原发性腹膜后肿瘤的全面放射学回顾

IF 1.8 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Clinical Imaging

Pub Date : 2024-11-02 DOI: 10.1016/j.clinimag.2024.110340

Yagmur Basak Polat, Mehmet Ali Gultekin, Ahmet Akcay, Ummuhan Ebru Karabulut, Bahar Atasoy, Huseyin Toprak

Primary retroperitoneal neoplasms (PRNs) are a complex and diverse group of tumors arising in the retroperitoneal space, excluding those from retroperitoneal organs. These masses present significant diagnostic challenges due to their heterogeneous nature. PRNs primarily include sarcomas, neurogenic tumors, extragonadal germ cell tumors, and lymphomas, with the majority being malignant. This necessitates thorough evaluation by radiologists to assess resectability and the need for biopsy.

Liposarcomas, the most common primary retroperitoneal sarcomas, and leiomyosarcomas, known for potential vessel involvement, exhibit distinct imaging patterns aiding differentiation. Neurogenic tumors, originating from nerve sheath, ganglionic, or paraganglionic cells, often appear in younger patients and have characteristic imaging features. Primary retroperitoneal extragonadal germ cell tumors are rare and are believed to originate from primordial germ cells that do not successfully migrate during embryonic development. Lymphomas are generally homogeneous on cross-sectional imaging; however, non-Hodgkin lymphomas can sometimes appear heterogeneous, complicating differentiation from other non-lipomatous retroperitoneal masses. Additionally, conditions like retroperitoneal fibrosis and Erdheim-Chester disease can mimic PRNs, complicating diagnosis and management.

This review aims to provide radiologists with essential diagnostic points for identifying PRNs, emphasizing the importance of precise imaging interpretation. Understanding these distinctions is vital for guiding clinical management and optimizing patient outcomes.

原发性腹膜后肿瘤（PRNs）是一组复杂多样的腹膜后肿瘤，不包括来自腹膜后器官的肿瘤。由于其异质性，这些肿块给诊断带来了巨大挑战。腹膜后肿瘤主要包括肉瘤、神经源性肿瘤、对角线外生殖细胞瘤和淋巴瘤，其中大多数为恶性。脂肪肉瘤是最常见的原发性腹膜后肉瘤，而子宫肌瘤则以潜在的血管受累而闻名，它们表现出不同的成像模式，有助于鉴别。神经源性肿瘤起源于神经鞘、神经节或副神经节细胞，通常出现在年轻患者身上，具有特征性的影像学特征。原发性腹膜后对角外生殖细胞瘤比较罕见，据信起源于胚胎发育过程中未能成功迁移的原始生殖细胞。在横断面成像上，淋巴瘤通常是均质的；但是，非霍奇金淋巴瘤有时也会出现异质性，这使得与其他非脂肪瘤性腹膜后肿块的鉴别变得复杂。此外，腹膜后纤维化和埃尔德海姆-切斯特病等疾病也可能与 PRNs 相似，从而使诊断和处理复杂化。本综述旨在为放射科医生提供鉴别 PRNs 的基本诊断要点，强调精确成像解读的重要性。了解这些区别对于指导临床治疗和优化患者预后至关重要。

{"title":"Beyond the surface: A comprehensive radiological review of primary retroperitoneal neoplasms","authors":"Yagmur Basak Polat, Mehmet Ali Gultekin, Ahmet Akcay, Ummuhan Ebru Karabulut, Bahar Atasoy, Huseyin Toprak","doi":"10.1016/j.clinimag.2024.110340","DOIUrl":"10.1016/j.clinimag.2024.110340","url":null,"abstract":"<div><div>Primary retroperitoneal neoplasms (PRNs) are a complex and diverse group of tumors arising in the retroperitoneal space, excluding those from retroperitoneal organs. These masses present significant diagnostic challenges due to their heterogeneous nature. PRNs primarily include sarcomas, neurogenic tumors, extragonadal germ cell tumors, and lymphomas, with the majority being malignant. This necessitates thorough evaluation by radiologists to assess resectability and the need for biopsy.</div><div>Liposarcomas, the most common primary retroperitoneal sarcomas, and leiomyosarcomas, known for potential vessel involvement, exhibit distinct imaging patterns aiding differentiation. Neurogenic tumors, originating from nerve sheath, ganglionic, or paraganglionic cells, often appear in younger patients and have characteristic imaging features. Primary retroperitoneal extragonadal germ cell tumors are rare and are believed to originate from primordial germ cells that do not successfully migrate during embryonic development. Lymphomas are generally homogeneous on cross-sectional imaging; however, non-Hodgkin lymphomas can sometimes appear heterogeneous, complicating differentiation from other non-lipomatous retroperitoneal masses. Additionally, conditions like retroperitoneal fibrosis and Erdheim-Chester disease can mimic PRNs, complicating diagnosis and management.</div><div>This review aims to provide radiologists with essential diagnostic points for identifying PRNs, emphasizing the importance of precise imaging interpretation. Understanding these distinctions is vital for guiding clinical management and optimizing patient outcomes.</div></div>","PeriodicalId":50680,"journal":{"name":"Clinical Imaging","volume":"116 ","pages":"Article 110340"},"PeriodicalIF":1.8,"publicationDate":"2024-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142592660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0