RESUMO
OBJECTIVE: The aim of this study was a prospective validation of the recently established ISGPS pancreas classification as a parenchymal risk classification system for pancreatic fistula after pancreatoduodenectomy. SUMMARY BACKGROUND DATA: Postoperative pancreatic fistula (POPF) is the major driver for complications after partial pancreatoduodenectomy (PD). Recently, the International Study Group for Pancreatic Surgery (ISGPS) published a pancreas classification containing the parameters main pancreatic duct diameter (MPD) and pancreatic texture to help assess the risk of POPF development following pancreatoduodenectomy. METHODS: From January 2020 to July 2021, 271 patients receiving elective PD were included after informed consent. The postoperative course was documented prospectively up to postoperative day 30. Among the pancreas characteristics, MPD and pancreatic texture were assessed intraoperatively at the pancreatic resection margin and the pancreatic glands were assigned to one of the four pancreas classes according to the ISGPS (A to D). The primary endpoint was POPF according to the updated ISGPS definition. Secondary endpoints comprised other post-PD morbidity and mortality. RESULTS: Of 271 patients, 264 had available data according to the ISGPS pancreas classification. Of those, 78 were assigned to class A (30%), 53 to class B (20%), 50 to class C (19%) and 83 to class D (31%). POPF occurred in 54 of 271 patients (19.9%). The 30-day mortality was 7/271 (2.6%), with 6/7 having developed POPF (86%). POPF rates within the classes A, B, C and D were 9.0%, 11.3%, 20.0% and 37.4%, respectively (P<0.001). In the univariable regression analysis, only patients in pancreas class D demonstrated a significantly higher risk for POPF when compared to class A (OR 6.05, 95%-CI: 2.6-15.9, P<0.001). In the multivariable regression model, patients in class D had a significantly higher risk for POPF compared to class A (OR 3.45, 95%-CI: 1.15-11.3, P=0.032). The model comprised Body Mass Index, surgery duration, microscopic fibrosis and the ISGPS pancreas classification, demonstrating an AUC-value of approximately 0.82 when tested on the PARIS dataset. CONCLUSION: This prospective trial shows that the ISGPS pancreas classification is valid. Patients in risk class D are prone to POPF independently of other factors. Therefore, all future publications on pancreatic surgery should report the risk class according to the ISGPS pancreas classification to allow for a better comparison of reported cohorts.
RESUMO
PURPOSE: The prevalent coronavirus disease 2019 (COVID-19) pandemic has spread throughout the world and is considered a serious threat to global health. The prognostic role of thoracic lymphadenopathy in COVID-19 is unclear. The aim of the present meta-analysis was to analyze the prognostic role of thoracic lymphadenopathy for the prediction of 30-day mortality in patients with COVID-19. MATERIALS AND METHODS: The MEDLINE library, Cochrane, and SCOPUS databases were screened for associations between CT-defined features and mortality in COVID-19 patients up to June 2021. In total, 21 studies were included in the present analysis. The quality of the included studies was assessed by the Newcastle-Ottawa Scale. The meta-analysis was performed using RevMan 5.3. Heterogeneity was calculated by means of the inconsistency index I2. DerSimonian and Laird random-effect models with inverse variance weights were performed without any further correction. RESULTS: The included studies comprised 4621 patients. The prevalence of thoracic lymphadenopathy varied between 1â% and 73.4â%. The pooled prevalence was 16.7â%, 95â% CIâ=â(15.6â%; 17.8â%). The hospital mortality was higher in patients with thoracic lymphadenopathy (34.7â%) than in patients without (20.0â%). The pooled odds ratio for the influence of thoracic lymphadenopathy on mortality was 2.13 (95â% CIâ=â[1.80-2.52], pâ<â0.001). CONCLUSION: The prevalence of thoracic lymphadenopathy in COVID-19 is 16.7â%. The presence of thoracic lymphadenopathy is associated with an approximately twofold increase in the risk for hospital mortality in COVID-19. KEY POINTS: · The prevalence of lymphadenopathy in COVID-19 is 16.7â%.. · Patients with lymphadenopathy in COVID-19 have a higher risk of mortality during hospitalization.. · Lymphadenopathy nearly doubles mortality and plays an important prognostic role.. CITATION FORMAT: · Bucher AM, Sieren M, Meinel F etâal. Prevalence and prognostic role of thoracic lymphadenopathy in Covid-19. Fortschr Röntgenstr 2024; DOI: 10.1055/a-2293-8132.
RESUMO
PURPOSE: To assess the feasibility and diagnostic accuracy of MRI-derived 3D volumetry of lower lumbar vertebrae and dural sac segments using shape-based machine learning for the detection of Marfan syndrome (MFS) compared with dural sac diameter ratios (the current clinical standard). MATERIALS AND METHODS: The final study sample was 144 patients being evaluated for MFS from 01/2012 to 12/2016, of whom 81 were non-MFS patients (46 [67%] female, 36 ± 16 years) and 63 were MFS patients (36 [57%] female, 35 ± 11 years) according to the 2010 Revised Ghent Nosology. All patients underwent 1.5T MRI with isotropic 1 × 1 × 1 mm3 3D T2-weighted acquisition of the lumbosacral spine. Segmentation and quantification of vertebral bodies L3-L5 and dural sac segments L3-S1 were performed using a shape-based machine learning algorithm. For comparison with the current clinical standard, anteroposterior diameters of vertebral bodies and dural sac were measured. Ratios between dural sac volume/diameter at the respective level and vertebral body volume/diameter were calculated. RESULTS: Three-dimensional volumetry revealed larger dural sac volumes (p < 0.001) and volume ratios (p < 0.001) at L3-S1 levels in MFS patients compared with non-MFS patients. For the detection of MFS, 3D volumetry achieved higher AUCs at L3-S1 levels (0.743, 0.752, 0.808, and 0.824) compared with dural sac diameter ratios (0.673, 0.707, 0.791, and 0.848); a significant difference was observed only for L3 (p < 0.001). CONCLUSION: MRI-derived 3D volumetry of the lumbosacral dural sac and vertebral bodies is a feasible method for quantifying dural ectasia using shape-based machine learning. Non-inferior diagnostic accuracy was observed compared with dural sac diameter ratio (the current clinical standard for MFS detection).
RESUMO
OBJECTIVES: To examine the effects of acupuncture and therapeutic exercise alone and in combination on temporomandibular joint symptoms in tension-type headache and to evaluate the potential interaction of existing temporomandibular dysfunction on the success of headache treatment. DESIGN: Pre-planned secondary analysis of a randomized controlled, non-blinded trial. SETTING: Outpatient clinic of a German university hospital. SUBJECTS: Ninety-six Participants with frequent episodic or chronic tension-type headache were randomized to one of four treatment groups. INTERVENTIONS: Six weeks of acupuncture or therapeutic exercise either as monotherapies or in combination, or usual care. Follow-up at 3 and 6 months. MAIN MEASURES: Subjective temporomandibular dysfunction symptoms were measured using the Functional Questionnaire Masticatory Organ, and the influence of this sum score and objective initial dental examination on the efficacy of headache treatment interventions was analyzed. RESULTS: Temporomandibular dysfunction score improved in all intervention groups at 3-month follow-up (usual care: 0.05 [SD 1.435]; acupuncture: -5 [SD 1.436]; therapeutic exercise: -4 [SD 1.798]; combination: -3 [SD 1.504]; P = 0.03). After 6 months, only acupuncture (-6 [SD 1.736]) showed a significant improvement compared to the usual care group (P < 0.01). Subjective temporomandibular dysfunction symptoms had no overall influence on headache treatment. CONCLUSIONS: Only acupuncture had long-lasting positive effects on the symptoms of temporomandibular dysfunction. Significant dental findings seem to inhibit the efficacy of acupuncture for tension-type headache.
Assuntos
Terapia por Acupuntura , Transtornos da Articulação Temporomandibular , Cefaleia do Tipo Tensional , Humanos , Cefaleia do Tipo Tensional/diagnóstico , Cefaleia do Tipo Tensional/etiologia , Cefaleia do Tipo Tensional/terapia , Terapia por Exercício , Cefaleia , Transtornos da Articulação Temporomandibular/terapia , Resultado do TratamentoRESUMO
PURPOSE: To assess the diagnostic accuracy of BMI-adapted, low-radiation and low-iodine dose, dual-source aortic CT for endoleak detection in non-obese and obese patients following endovascular aortic repair. METHODS: In this prospective single-center study, patients referred for follow-up CT after endovascular repair with a history of at least one standard triphasic (native, arterial and delayed phase) routine CT protocol were enrolled. Patients were divided into two groups and allocated to a BMI-adapted (group A, BMI < 30 kg/m2; group B, BMI ≥ 30 kg/m2) double low-dose CT (DLCT) protocol comprising single-energy arterial and dual-energy delayed phase series with virtual non-contrast (VNC) reconstructions. An in-patient comparison of the DLCT and routine CT protocol as reference standard was performed regarding differences in diagnostic accuracy, radiation dose, and image quality. RESULTS: Seventy-five patients were included in the study (mean age 73 ± 8 years, 63 (84%) male). Endoleaks were diagnosed in 20 (26.7%) patients, 11 of 53 (20.8%) in group A and 9 of 22 (40.9%) in group B. Two radiologists achieved an overall diagnostic accuracy of 98.7% and 97.3% for endoleak detection, with 100% in group A and 95.5% and 90.9% in group B. All examinations were diagnostic. The DLCT protocol reduced the effective dose from 10.0 ± 3.6 mSv to 6.1 ± 1.5 mSv (p < 0.001) and the total iodine dose from 31.5 g to 14.5 g in group A and to 17.4 g in group B. CONCLUSION: Optimized double low-dose dual-source aortic CT with VNC, arterial and delayed phase images demonstrated high diagnostic accuracy for endoleak detection and significant radiation and iodine dose reductions in both obese and non-obese patients compared to the reference standard of triple phase, standard radiation and iodine dose aortic CT.
RESUMO
RATIONALE AND OBJECTIVES: The prognostic role of pericardial effusion (PE) in Covid 19 is unclear. The aim of the present study was to estimate the prognostic role of PE in patients with Covid 19 in a large multicentre setting. MATERIALS AND METHODS: This retrospective study is a part of the German multicenter project RACOON (Radiological Cooperative Network of the Covid 19 pandemic). The acquired sample comprises 1197 patients, 363 (30.3%) women and 834 (69.7%) men. In every case, chest computed tomography was analyzed for PE. Data about 30-day mortality, need for mechanical ventilation and need for intensive care unit (ICU) admission were collected. Data were evaluated by means of descriptive statistics. Group differences were calculated with Mann-Whitney test and Fisher exact test. Uni-and multivariable regression analyses were performed. RESULTS: Overall, 46.4% of the patients were admitted to ICU, mechanical lung ventilation was performed in 26.6% and 30-day mortality was 24%. PE was identified in 159 patients (13.3%). The presence of PE was associated with 30-day mortality: HR= 1.54, CI 95% (1.05; 2.23), p = 0.02 (univariable analysis), and HR= 1.60, CI 95% (1.03; 2.48), p = 0.03 (multivariable analysis). Furthermore, density of PE was associated with the need for intubation (OR=1.02, CI 95% (1.003; 1.05), p = 0.03) and the need for ICU admission (OR=1.03, CI 95% (1.005; 1.05), p = 0.01) in univariable regression analysis. The presence of PE was associated with 30-day mortality in male patients, HR= 1.56, CI 95%(1.01-2.43), p = 0.04 (multivariable analysis). In female patients, none of PE values predicted clinical outcomes. CONCLUSION: The prevalence of PE in Covid 19 is 13.3%. PE is an independent predictor of 30-day mortality in male patients with Covid 19. In female patients, PE plays no predictive role.
Assuntos
COVID-19 , Derrame Pericárdico , Tomografia Computadorizada por Raios X , Humanos , Masculino , Feminino , COVID-19/mortalidade , COVID-19/epidemiologia , COVID-19/diagnóstico por imagem , COVID-19/complicações , Estudos Retrospectivos , Derrame Pericárdico/diagnóstico por imagem , Derrame Pericárdico/epidemiologia , Idoso , Pessoa de Meia-Idade , Prognóstico , Alemanha/epidemiologia , Respiração Artificial/estatística & dados numéricos , SARS-CoV-2 , Unidades de Terapia Intensiva , Idoso de 80 Anos ou maisRESUMO
The Lon protease is a highly conserved protein degradation machine that has critical regulatory and protein quality control functions in cells from the three domains of life. Here, we report the discovery of a α-proteobacterial heat shock protein, LarA, that functions as a dedicated Lon regulator. We show that LarA accumulates at the onset of proteotoxic stress and allosterically activates Lon-catalysed degradation of a large group of substrates through a five amino acid sequence at its C-terminus. Further, we find that high levels of LarA cause growth inhibition in a Lon-dependent manner and that Lon-mediated degradation of LarA itself ensures low LarA levels in the absence of stress. We suggest that the temporal LarA-dependent activation of Lon helps to meet an increased proteolysis demand in response to protein unfolding stress. Our study defines a regulatory interaction of a conserved protease with a heat shock protein, serving as a paradigm of how protease activity can be tuned under changing environmental conditions.
Assuntos
Proteínas de Escherichia coli , Protease La , Protease La/genética , Protease La/metabolismo , Proteínas de Choque Térmico/genética , Proteínas de Choque Térmico/metabolismo , Proteínas de Escherichia coli/metabolismo , Estresse Proteotóxico , Endopeptidases/metabolismo , Proteases Dependentes de ATP/metabolismoRESUMO
IMPORTANCE: Regulated protein degradation is a critical process in all cell types, which contributes to the precise regulation of protein amounts in response to internal and external cues. In bacteria, protein degradation is carried out by ATP-dependent proteases. Although past work revealed detailed insights into the operation principles of these proteases, there is limited knowledge about the substrate proteins that are degraded by distinct proteases and the regulatory role of proteolysis in cellular processes. This study reveals a direct role of the conserved protease Lon in regulating σT, a transcriptional regulator of the general stress response in α-proteobacteria. Our work is significant as it underscores the importance of regulated proteolysis in modulating the levels of key regulatory proteins under changing conditions.
Assuntos
Caulobacter crescentus , Protease La , Proteólise , Proteínas de Bactérias/genética , Proteínas de Bactérias/metabolismo , Caulobacter crescentus/metabolismo , Regulação Bacteriana da Expressão Gênica , Protease La/genética , Protease La/metabolismo , Fator sigma/genética , Fator sigma/metabolismoAssuntos
Processamento de Linguagem Natural , Radiologia , Humanos , Radiografia , Relatório de PesquisaRESUMO
Background The latest large language models (LLMs) solve unseen problems via user-defined text prompts without the need for retraining, offering potentially more efficient information extraction from free-text medical records than manual annotation. Purpose To compare the performance of the LLMs ChatGPT and GPT-4 in data mining and labeling oncologic phenotypes from free-text CT reports on lung cancer by using user-defined prompts. Materials and Methods This retrospective study included patients who underwent lung cancer follow-up CT between September 2021 and March 2023. A subset of 25 reports was reserved for prompt engineering to instruct the LLMs in extracting lesion diameters, labeling metastatic disease, and assessing oncologic progression. This output was fed into a rule-based natural language processing pipeline to match ground truth annotations from four radiologists and derive performance metrics. The oncologic reasoning of LLMs was rated on a five-point Likert scale for factual correctness and accuracy. The occurrence of confabulations was recorded. Statistical analyses included Wilcoxon signed rank and McNemar tests. Results On 424 CT reports from 424 patients (mean age, 65 years ± 11 [SD]; 265 male), GPT-4 outperformed ChatGPT in extracting lesion parameters (98.6% vs 84.0%, P < .001), resulting in 96% correctly mined reports (vs 67% for ChatGPT, P < .001). GPT-4 achieved higher accuracy in identification of metastatic disease (98.1% [95% CI: 97.7, 98.5] vs 90.3% [95% CI: 89.4, 91.0]) and higher performance in generating correct labels for oncologic progression (F1 score, 0.96 [95% CI: 0.94, 0.98] vs 0.91 [95% CI: 0.89, 0.94]) (both P < .001). In oncologic reasoning, GPT-4 had higher Likert scale scores for factual correctness (4.3 vs 3.9) and accuracy (4.4 vs 3.3), with a lower rate of confabulation (1.7% vs 13.7%) than ChatGPT (all P < .001). Conclusion When using user-defined prompts, GPT-4 outperformed ChatGPT in extracting oncologic phenotypes from free-text CT reports on lung cancer and demonstrated better oncologic reasoning with fewer confabulations. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Hafezi-Nejad and Trivedi in this issue.
Assuntos
Neoplasias Pulmonares , Segunda Neoplasia Primária , Humanos , Masculino , Idoso , Estudos Retrospectivos , Neoplasias Pulmonares/diagnóstico por imagem , Mineração de Dados , Oncologia , Benchmarking , Transtornos da MemóriaRESUMO
BACKGROUND: With the introduction of ChatGPT in late November 2022, large language models based on artificial intelligence have gained worldwide recognition. These language models are trained on vast amounts of data, enabling them to process complex tasks in seconds and provide detailed, high-level text-based responses. OBJECTIVE: To provide an overview of the most widely discussed large language models, ChatGPT and GPT4, with a focus on potential applications for patient-centered radiology. MATERIALS AND METHODS: A PubMed search of both large language models was performed using the terms "ChatGPT" and "GPT-4", with subjective selection and completion in the form of a narrative review. RESULTS: The generic nature of language models holds great promise for radiology, enabling both patients and referrers to facilitate understanding of radiological findings, overcome language barriers, and improve the quality of informed consent discussions. This could represent a significant step towards patient-centered or person-centered radiology. CONCLUSION: Large language models represent a promising tool for improving the communication of findings, interdisciplinary collaboration, and workflow in radiology. However, important privacy issues and the reliable applicability of these models in medicine remain to be addressed.
Assuntos
Inteligência Artificial , Radiologia , Humanos , Radiografia , Idioma , Assistência Centrada no PacienteRESUMO
With the implementation of the new MDR 2017/745 by the European Parliament, more robust clinical and pre-clinical data will be required due to a more stringent approval process. The EFORT Implant and Patient Safety Initiative WG1 'Introduction of Innovation', combined knowledge of orthopaedic surgeons, research institutes, orthopaedic device manufacturers, patient representatives and regulatory authorities to develop a comprehensive set of recommendations for the introduction of innovations in joint arthroplasty within the boundaries of MDR 2017/745. Recommendations have been developed to address key questions about pre-clinical and clinical requirements for the introduction of new implants and implant-related instrumentation with the participation of a steering group, invited by the EFORT Board in dialogue with representatives from European National Societies and Speciality Societies. Different degrees of novelty and innovation were described and agreed on in relation to when surgeons can start, using implants and implant-related instrumentation routinely. Before any clinical phase of a new implant, following the pre-market clinical investigation or the equivalent device PMCF pathway, it is a common understanding that all appropriate pre-clinical testing (regulatory mandatory and evident state of the art) - which has to be considered for a specific device - has been successfully completed. Once manufacturers receive the CE mark for a medical device, it can be used in patients routinely when a clinical investigation has been conducted to demonstrate the conformity of devices according to MDR Article 62 or full equivalence for the technical, biological and clinical characteristics has been demonstrated (MDR, Annex XIV, Part A, 3.) and a PMCF study has been initiated.
RESUMO
In May 2021, the new Medical Device Regulation in the EU came into force. While the US has a centralized governmental authority, the Food and Drug Administration (FDA), the EU implemented a system of different Notified Bodies responsible for the approval process of medical devices. Both regions have a similar system to classify medical devices based on their overall risks but specific devices, like joint prostheses, are classified differently in the US and the EU. Depending on the risk class, there are differences in the quality and quantity of clinical data required to obtain market approval. In both regions, it is possible to place a new device on the market based on the demonstration of equivalence to an already marketed device, but the MDR significantly increased the regulatory requirements for the equivalence pathway. While an approved medical device in the US in most cases only requires general post-market surveillance activities, manufacturers in the EU must continuously collect clinical data and submit specific reports to the Notified Bodies. In this article, we will compare the regulatory requirements between the US and Europe and provide an overview of similarities and differences.
Assuntos
Legislação de Dispositivos Médicos , Estados Unidos , Humanos , Europa (Continente) , United States Food and Drug AdministrationRESUMO
OBJECTIVES: To compare the effects of acupuncture and medical training therapy in combination or individually with usual care on quality of life, depression, and anxiety in patients with tension-type headache. METHODS: In this single-center, prospective, randomized, controlled, unblinded trial, 96 adults (38.7(+/-13.3) years of age; 75 females/20 males/one dropout) with frequent episodic or chronic tension-type headache were randomized to one of four treatment groups (n = 24). The treatment groups received six weeks of either acupuncture or medical training therapy as monotherapies or in combination (12 interventions each), or usual care. We assessed depressiveness (PHQ-9), anxiety (GAD-7), and health-related quality of life (SF-12) as secondary outcome parameters at baseline, six weeks, three months, and six months after initiation of treatment. Linear mixed models were calculated. RESULTS: Both, acupuncture (baseline to six-weeks change scores: mean: -2(standard deviation: 2.5 points), three months: -2.4(2.4), six-months -2.7(3.6)) and the combination of acupuncture and medical training therapy (-2.7(4.9), -2.2(4.0), -2.2(4.2)) (each within-group p < .05) significantly reduced depressiveness-scores (PHQ-9) to a greater extent than medical training therapy (-0.3(2.0), -0.5(1.6), -0.9(2.6)) or usual care alone (-0.8(2.9), 0.1(2.8), 0.2(3.6)). We found similar results with anxiety scores and the physical sum scores of the SF-12. No severe adverse events occurred. CONCLUSIONS: Acupuncture and the combination of acupuncture and medical training therapy elicit positive effects on depression, anxiety, quality of life, and symptom intensity in patients with episodic and chronic tension-type headache. Acupuncture appears to play a central role in mediating the therapeutic effects, underscoring the clinical relevance of this treatment. An additive benefit of the combination of both therapies does not appear to be relevant.Trial registration: Registered on 11 February 2019. German Clinical Trials Register, DRKS00016723.
Assuntos
Terapia por Acupuntura , Cefaleia do Tipo Tensional , Adulto , Feminino , Humanos , Masculino , Terapia por Acupuntura/métodos , Ansiedade/terapia , Depressão/terapia , Estudos Prospectivos , Qualidade de Vida , Cefaleia do Tipo Tensional/diagnóstico , Pessoa de Meia-IdadeRESUMO
The topics of ethical conduct and governance in academic research in the business field have attracted scientific and public attention. The concern is that research misconduct in organizations such as business schools and universities might result in practitioners, policymakers, and researchers grounding their decisions on biased research results. This study addresses ethical research misconduct by investigating whether the ethical orientation of business researchers is related to the likelihood of research misconduct, such as selective reporting of research findings. We distinguish between deontological and consequentialist ethical orientations and the competition between researchers and investigate the moderating role of their perceived autonomy. Based on global data collected from 1031 business scholars, we find that researchers with a strong deontological ethical orientation are less prone to misconduct. This effect is robust against different levels of perceived autonomy and competition. In contrast, researchers having a consequentialist ethical orientation is positively associated with misconduct in business research. High levels of competition in the research environment reinforce this effect. Our results reveal a potentially toxic combination comprising researchers with a strong consequentialist orientation who are embedded in highly competitive research environments. Our research calls for the development of ethical orientations grounded on maxims rather than anticipated consequences among researchers. We conclude that measures for ethical governance in business schools should consider the ethical orientation that underlies researchers' decision-making and the organizational and institutional environment in which business researchers are embedded.
RESUMO
Obesity-related metabolic disorders such as hypertension, hyperlipidemia and chronic inflammation have been associated with aortic dilatation and resulting in aortic aneurysms in many cases. Whether weight loss may reduce the risk of aortic dilatation is not clear. In this study, the diameter of the descending thoracic aorta, infrarenal abdominal aorta and aortic bifurcation of 144 overweight or obese non-smoking adults were measured by MR-imaging, at baseline, and 12 and 50 weeks after weight loss by calorie restriction. Changes in aortic diameter, anthropometric measures and body composition and metabolic markers were evaluated using linear mixed models. The association of the aortic diameters with the aforementioned clinical parameters was analyzed using Spearman`s correlation. Weight loss was associated with a reduction in the thoracic and abdominal aortic diameters 12 weeks after weight loss (predicted relative differences for Quartile 4: 2.5% ± 0.5 and -2.2% ± 0.8, p < 0.031; respectively). Furthermore, there was a nominal reduction in aortic diameters during the 50-weeks follow-up period. Aortic diameters were positively associated with weight, visceral adipose tissue, glucose, HbA1c and with both systolic and diastolic blood pressure. Weight loss induced by calorie restriction may reduce aortic diameters. Future studies are needed to investigate, whether the reduction of aortic diameters via calorie restriction may help to prevent aortic aneurysms.
RESUMO
Purpose: To train a deep natural language processing (NLP) model, using data mined structured oncology reports (SOR), for rapid tumor response category (TRC) classification from free-text oncology reports (FTOR) and to compare its performance with human readers and conventional NLP algorithms. Materials and Methods: In this retrospective study, databases of three independent radiology departments were queried for SOR and FTOR dated from March 2018 to August 2021. An automated data mining and curation pipeline was developed to extract Response Evaluation Criteria in Solid Tumors-related TRCs for SOR for ground truth definition. The deep NLP bidirectional encoder representations from transformers (BERT) model and three feature-rich algorithms were trained on SOR to predict TRCs in FTOR. Models' F1 scores were compared against scores of radiologists, medical students, and radiology technologist students. Lexical and semantic analyses were conducted to investigate human and model performance on FTOR. Results: Oncologic findings and TRCs were accurately mined from 9653 of 12 833 (75.2%) queried SOR, yielding oncology reports from 10 455 patients (mean age, 60 years ± 14 [SD]; 5303 women) who met inclusion criteria. On 802 FTOR in the test set, BERT achieved better TRC classification results (F1, 0.70; 95% CI: 0.68, 0.73) than the best-performing reference linear support vector classifier (F1, 0.63; 95% CI: 0.61, 0.66) and technologist students (F1, 0.65; 95% CI: 0.63, 0.67), had similar performance to medical students (F1, 0.73; 95% CI: 0.72, 0.75), but was inferior to radiologists (F1, 0.79; 95% CI: 0.78, 0.81). Lexical complexity and semantic ambiguities in FTOR influenced human and model performance, revealing maximum F1 score drops of -0.17 and -0.19, respectively. Conclusion: The developed deep NLP model reached the performance level of medical students but not radiologists in curating oncologic outcomes from radiology FTOR.Keywords: Neural Networks, Computer Applications-Detection/Diagnosis, Oncology, Research Design, Staging, Tumor Response, Comparative Studies, Decision Analysis, Experimental Investigations, Observer Performance, Outcomes Analysis Supplemental material is available for this article. © RSNA, 2022.
RESUMO
Detector-based spectral CT offers the possibility of obtaining spectral information from which discrete acquisitions at different energy levels can be derived, yielding so-called virtual monoenergetic images (VMI). In this study, we aimed to develop a jointly optimized deep-learning framework based on dual-energy CT pulmonary angiography (DE-CTPA) data to generate synthetic monoenergetic images (SMI) for improving automatic pulmonary embolism (PE) detection in single-energy CTPA scans. For this purpose, we used two datasets: our institutional DE-CTPA dataset D1, comprising polyenergetic arterial series and the corresponding VMI at low-energy levels (40 keV) with 7892 image pairs, and a 10% subset of the 2020 RSNA Pulmonary Embolism CT Dataset D2, which consisted of 161,253 polyenergetic images with dichotomous slice-wise annotations (PE/no PE). We trained a fully convolutional encoder-decoder on D1 to generate SMI from single-energy CTPA scans of D2, which were then fed into a ResNet50 network for training of the downstream PE classification task. The quantitative results on the reconstruction ability of our framework revealed high-quality visual SMI predictions with reconstruction results of 0.984 ± 0.002 (structural similarity) and 41.706 ± 0.547 dB (peak signal-to-noise ratio). PE classification resulted in an AUC of 0.84 for our model, which achieved improved performance compared to other naïve approaches with AUCs up to 0.81. Our study stresses the role of using joint optimization strategies for deep-learning algorithms to improve automatic PE detection. The proposed pipeline may prove to be beneficial for computer-aided detection systems and could help rescue CTPA studies with suboptimal opacification of the pulmonary arteries from single-energy CT scanners.
RESUMO
OBJECTIVE: To develop an automatic COVID-19 Reporting and Data System (CO-RADS)-based classification in a multi-demographic setting. METHODS: This multi-institutional review boards-approved retrospective study included 2720 chest CT scans (mean age, 58 years [range 18-100 years]) from Italian and Russian patients. Three board-certified radiologists from three countries assessed randomly selected subcohorts from each population and provided CO-RADS-based annotations. CT radiomic features were extracted from the selected subcohorts after preprocessing steps like lung lobe segmentation and automatic noise reduction. We compared three machine learning models, logistic regression (LR), multilayer perceptron (MLP), and random forest (RF) for the automated CO-RADS classification. Model evaluation was carried out in two scenarios, first, training on a mixed multi-demographic subcohort and testing on an independent hold-out dataset. In the second scenario, training was done on a single demography and externally validated on the other demography. RESULTS: The overall inter-observer agreement for the CO-RADS scoring between the radiologists was substantial (k = 0.80). Irrespective of the type of validation test scenario, suspected COVID-19 CT scans were identified with an accuracy of 84%. SHapley Additive exPlanations (SHAP) interpretation showed that the "wavelet_(LH)_GLCM_Imc1" feature had a positive impact on COVID prediction both with and without noise reduction. The application of noise reduction improved the overall performance between the classifiers for all types. CONCLUSION: Using an automated model based on the COVID-19 Reporting and Data System (CO-RADS), we achieved clinically acceptable performance in a multi-demographic setting. This approach can serve as a standardized tool for automated COVID-19 assessment. KEYPOINTS: ⢠Automatic CO-RADS scoring of large-scale multi-demographic chest CTs with mean AUC of 0.93 ± 0.04. ⢠Validation procedure resembles TRIPOD 2b and 3 categories, enhancing the quality of experimental design to test the cross-dataset domain shift between institutions aiding clinical integration. ⢠Identification of COVID-19 pneumonia in the presence of community-acquired pneumonia and other comorbidities with an AUC of 0.92.
Assuntos
COVID-19 , Pneumonia , Adolescente , Adulto , Idoso , Idoso de 80 Anos ou mais , Demografia , Humanos , Pessoa de Meia-Idade , Estudos Retrospectivos , Tomografia Computadorizada por Raios X/métodos , Adulto JovemRESUMO
INTRODUCTION: Partial pancreatoduodenectomy (PD) is the treatment of choice for many malignant and benign diseases of the pancreatic head. Postoperative complication rates of up to 40% are regularly reported. One of the most common and potentially life-threatening complication is the postoperative pancreatic fistula (POPF). Parenchymal risk factors like main pancreatic duct diameter or texture of the pancreatic gland have already been identified in retrospective studies. The aim of this study is to evaluate the diagnostic value of parenchymal risk factors on POPF in a prospective manner. METHODS AND ANALYSIS: All patients scheduled for elective PD at the Department of General, Visceral and Transplantation Surgery of the University of Heidelberg will be screened for eligibility. As diagnostic factors, diameter and texture of the pancreatic gland as well as radiological and histopathological features will be recorded. Furthermore, the new four class risk classification system by the International Study Group of Pancreatic Surgery (ISGPS) will be recorded. The postoperative course will be monitored prospectively. The primary endpoint will be the association of the main pancreatic duct size and the texture of the pancreatic gland on POPF according to the updated ISGPS definition. The diagnostic value of the above-mentioned factors for POPF will be evaluated in a univariable and multivariable analysis. ETHICS AND DISSEMINATION: PARIS is a monocentric, prospective, diagnostic study to evaluate the association of parenchymal risk factors and the development of POPF approved by the Ethics Committee of the medical faculty of Heidelberg University (S-344/2019). Results will be available in 2022 and will be published at national and international meetings. With this knowledge, the intraoperative and perioperative decision-making process could be eased and improve the individual outcome of patient. TRIAL REGISTRATION NUMBER: DRKS00017184.