RESUMEN
BACKGROUND. Imaging reports that consistently document all disease sites with a potential to increase surgical complexity or morbidity can facilitate ovarian cancer treatment planning. OBJECTIVE. The aims of this study were to compare simple structured reports and synoptic reports from pretreatment CT examinations in patients with advanced ovarian cancer in terms of completeness of documenting involvement of clinically relevant anatomic sites as well as to evaluate physician satisfaction with synoptic reports. METHODS. This retrospective study included 205 patients (median age, 65 years) who underwent contrast-enhanced abdominopelvic CT before primary treatment of advanced ovarian cancer from June 1, 2018, to January 31, 2022. A total of 128 reports generated on or before March 31, 2020, used a simple structured report (free text organized into sections); 77 reports generated on or after April 1, 2020, used a synoptic report (a list of 45 anatomic sites relevant to ovarian cancer management, each of which was classified in terms of disease absence versus presence). Reports were reviewed for completeness of documentation of involvement of the 45 sites. For patients who underwent neoadjuvant chemotherapy based on diagnostic laparoscopy findings or underwent primary debulking surgery with suboptimal resection, the EMR was reviewed to identify surgically established sites of disease that were unresectable or challenging to resect. Gynecologic oncology surgeons were electronically surveyed. RESULTS. The mean report turnaround time was 29.8 minutes for simple structured reports versus 54.5 minutes for synoptic reports (p < .001). A mean of 17.6 of 45 sites (range, four to 43 sites) were mentioned by simple structured reports versus 44.5 of 45 sites (range, 39-45) for synoptic reports (p < .001). Forty-three patients had surgically established unresectable or challenging-to-resect disease; involvement of anatomic site(s) with such disease was mentioned in 37% (11/30) of simple structured reports versus 100% (13/13) of synoptic reports (p < .001). All eight surveyed gynecologic oncology surgeons completed the survey. CONCLUSION. A synoptic report improved completeness of pretreatment CT reports in patients with advanced ovarian cancer, including for established sites of unresectable or challenging-to-resect disease. CLINICAL IMPACT. The findings indicate the role of disease-specific synoptic reports in facilitating referrer communication and potentially guiding clinical decision-making.
Asunto(s)
Neoplasias de los Genitales Femeninos , Neoplasias Ováricas , Médicos , Humanos , Femenino , Anciano , Estudios Retrospectivos , Satisfacción del Paciente , Neoplasias Ováricas/diagnóstico por imagen , Neoplasias Ováricas/cirugía , Documentación , Tomografía Computarizada por Rayos X , Satisfacción PersonalRESUMEN
OBJECTIVES: While fully supervised learning can yield high-performing segmentation models, the effort required to manually segment large training sets limits practical utility. We investigate whether data mined line annotations can facilitate brain MRI tumor segmentation model development without requiring manually segmented training data. METHODS: In this retrospective study, a tumor detection model trained using clinical line annotations mined from PACS was leveraged with unsupervised segmentation to generate pseudo-masks of enhancing tumors on T1-weighted post-contrast images (9911 image slices; 3449 adult patients). Baseline segmentation models were trained and employed within a semi-supervised learning (SSL) framework to refine the pseudo-masks. Following each self-refinement cycle, a new model was trained and tested on a held-out set of 319 manually segmented image slices (93 adult patients), with the SSL cycles continuing until Dice score coefficient (DSC) peaked. DSCs were compared using bootstrap resampling. Utilizing the best-performing models, two inference methods were compared: (1) conventional full-image segmentation, and (2) a hybrid method augmenting full-image segmentation with detection plus image patch segmentation. RESULTS: Baseline segmentation models achieved DSC of 0.768 (U-Net), 0.831 (Mask R-CNN), and 0.838 (HRNet), improving with self-refinement to 0.798, 0.871, and 0.873 (each p < 0.001), respectively. Hybrid inference outperformed full image segmentation alone: DSC 0.884 (Mask R-CNN) vs. 0.873 (HRNet), p < 0.001. CONCLUSIONS: Line annotations mined from PACS can be harnessed within an automated pipeline to produce accurate brain MRI tumor segmentation models without manually segmented training data, providing a mechanism to rapidly establish tumor segmentation capabilities across radiology modalities. KEY POINTS: ⢠A brain MRI tumor detection model trained using clinical line measurement annotations mined from PACS was leveraged to automatically generate tumor segmentation pseudo-masks. ⢠An iterative self-refinement process automatically improved pseudo-mask quality, with the best-performing segmentation pipeline achieving a Dice score of 0.884 on a held-out test set. ⢠Tumor line measurement annotations generated in routine clinical radiology practice can be harnessed to develop high-performing segmentation models without manually segmented training data, providing a mechanism to rapidly establish tumor segmentation capabilities across radiology modalities.
Asunto(s)
Neoplasias Encefálicas , Procesamiento de Imagen Asistido por Computador , Adulto , Humanos , Procesamiento de Imagen Asistido por Computador/métodos , Estudios Retrospectivos , Imagen por Resonancia Magnética/métodos , Neoplasias Encefálicas/diagnóstico por imagen , Encéfalo/diagnóstico por imagenRESUMEN
CONTEXT: Health departments (HDs) work on the front lines to ensure the health of their communities, providing a unique perspective to public health response activities. Say Yes! COVID Test (SYCT) is a US federally funded program providing free COVID-19 self-tests to communities with high COVID-19 transmission, low vaccination rates, and high social vulnerability. The collaboration with 9 HDs was key for the program distribution of 5.8 million COVID-19 self-tests between March 31 and November 30, 2021. OBJECTIVE: The objective of this study was to gather qualitative in-depth information on the experiences of HDs with the SYCT program to better understand the successes and barriers to implementing community-focused self-testing programs. DESIGN: Key informant (KI) interviews. SETTING: Online interviews conducted between November and December 2021. PARTICIPANTS: Sixteen program leads representing 9 HDs were purposefully sampled as KIs. KIs completed 60-minute structured interviews conducted by one trained facilitator and recorded. MAIN OUTCOME MEASURES: Key themes and lessons learned were identified using grounded theory. RESULTS: Based on perceptions of KIs, HDs that maximized community partnerships for test distribution were more certain that populations at a higher risk for COVID-19 were reached. Where the HD relied predominantly on direct-to-consumer distribution, KIs were less certain that communities at higher risk were served. Privacy and anonymity in testing were themes linked to higher perceived community acceptance. KIs reported that self-test demand and distribution levels increased during higher COVID-19 transmission levels. CONCLUSION: HDs that build bridges and engage with community partners and trusted leaders are better prepared to identify and link high-risk populations with health services and resources. When collaborating with trusted community organizations, KIs perceived that the SYCT program overcame barriers such as mistrust of government intervention and desire for privacy and motivated community members to utilize this resource to protect themselves against COVID-19.
Asunto(s)
COVID-19 , Humanos , COVID-19/diagnóstico , COVID-19/epidemiología , COVID-19/prevención & control , Autoevaluación , Prueba de COVID-19 , Teoría Fundamentada , Salud PúblicaRESUMEN
A vascular lab procedure culminates in a diagnostic report that is a combination of the data generated on the vascular lab equipment, physician interpretations, and recommendations. The transcription process can be error prone and inefficient. Computerized capture of data from the equipment and transmission into a reporting system is the definition of "automation" in this article. In this article, we describe an organized approach to collecting data generated on vascular lab equipment and delivering to reporting systems to decrease error and improve efficiency.
Asunto(s)
Automatización , HumanosRESUMEN
Radiological reports are a valuable source of information used to guide clinical care and support research. Organizing and managing this content, however, frequently requires several manual curations because of the more common unstructured nature of the reports. However, manual review of these reports for clinical knowledge extraction is costly and time-consuming. Natural language processing (NLP) is a set of methods developed to extract structured meaning from a body of text and can be used to optimize the workflow of health care professionals. Specifically, NLP methods can help radiologists as decision support systems and improve the management of patients' medical data. In this study, we highlight the opportunities offered by NLP in the field of radiology. A comprehensive review of the most commonly used NLP methods to extract information from radiological reports and the development of tools to improve radiological workflow using this information is presented. Finally, we review the important limitations of these tools and discuss the relevant observations and trends in the application of NLP to radiology that could benefit the field in the future.
Asunto(s)
Procesamiento de Lenguaje Natural , Radiología , Humanos , Radiografía , Radiólogos , Informe de InvestigaciónRESUMEN
PURPOSE: To determine how particle density affects dose distribution and outcomes after lobar radioembolization. METHODS: Matched pairs of patients, treated with glass versus resin microspheres, were selected by propensity score matching (114 patients), in this single-institution retrospective study. For each patient, tumor and liver particle density (particles/cm3) and dose (Gy) were determined. Tumor-to-normal ratio was measured on both 99mTc-MAA SPECT/CT and post-90Y bremsstrahlung SPECT/CT. Microdosimetry simulations were used to calculate first percentile dose, which is the dose in the cold spots between microspheres. Local progression-free survival (LPFS) and overall survival were analyzed. RESULTS: As more particles were delivered, doses on 90Y SPECT/CT became more uniform throughout the treatment volume: tumor and liver doses became more similar (p = 0.04), and microscopic cold spots between particles disappeared. For hypervascular tumors (tumor-to-normal ratio ≥ 2.6 on MAA scan), delivering fewer particles (< 6000 particles/cm3 treatment volume) was associated with better LPFS (p = 0.03). For less vascular tumors (tumor-to-normal ratio < 2.6), delivering more particles (≥ 6000 particles/cm3) was associated with better LPFS (p = 0.02). In matched pairs of patients, using the optimal particle density resulted in improved overall survival (11.5 vs. 6.8 months, p = 0.047), compared to using suboptimal particle density. Microdosimetry resulted in better predictions of LPFS (p = 0.03), and overall survival (p = 0.02), compared to conventional dosimetry. CONCLUSION: The number of particles delivered can be chosen to maximize the tumor dose and minimize the liver dose, based on tumor vascularity. Optimizing the particle density resulted in improved LPFS and overall survival.
Asunto(s)
Carcinoma Hepatocelular , Embolización Terapéutica , Neoplasias Hepáticas , Carcinoma Hepatocelular/diagnóstico por imagen , Carcinoma Hepatocelular/radioterapia , Embolización Terapéutica/métodos , Humanos , Neoplasias Hepáticas/diagnóstico por imagen , Neoplasias Hepáticas/radioterapia , Microesferas , Estudios Retrospectivos , Agregado de Albúmina Marcado con Tecnecio Tc 99m , Tomografía Computarizada de Emisión de Fotón Único , Radioisótopos de Itrio/uso terapéuticoRESUMEN
PURPOSE: To develop a deep network architecture that would achieve fully automated radiologist-level segmentation of cancers at breast MRI. MATERIALS AND METHODS: In this retrospective study, 38 229 examinations (composed of 64 063 individual breast scans from 14 475 patients) were performed in female patients (age range, 12-94 years; mean age, 52 years ± 10 [standard deviation]) who presented between 2002 and 2014 at a single clinical site. A total of 2555 breast cancers were selected that had been segmented on two-dimensional (2D) images by radiologists, as well as 60 108 benign breasts that served as examples of noncancerous tissue; all these were used for model training. For testing, an additional 250 breast cancers were segmented independently on 2D images by four radiologists. Authors selected among several three-dimensional (3D) deep convolutional neural network architectures, input modalities, and harmonization methods. The outcome measure was the Dice score for 2D segmentation, which was compared between the network and radiologists by using the Wilcoxon signed rank test and the two one-sided test procedure. RESULTS: The highest-performing network on the training set was a 3D U-Net with dynamic contrast-enhanced MRI as input and with intensity normalized for each examination. In the test set, the median Dice score of this network was 0.77 (interquartile range, 0.26). The performance of the network was equivalent to that of the radiologists (two one-sided test procedures with radiologist performance of 0.69-0.84 as equivalence bounds, P < .001 for both; n = 250). CONCLUSION: When trained on a sufficiently large dataset, the developed 3D U-Net performed as well as fellowship-trained radiologists in detailed 2D segmentation of breast cancers at routine clinical MRI.Keywords: MRI, Breast, Segmentation, Supervised Learning, Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning AlgorithmsPublished under a CC BY 4.0 license. Supplemental material is available for this article.
RESUMEN
Background Artificial intelligence (AI) applications for cancer imaging conceptually begin with automated tumor detection, which can provide the foundation for downstream AI tasks. However, supervised training requires many image annotations, and performing dedicated post hoc image labeling is burdensome and costly. Purpose To investigate whether clinically generated image annotations can be data mined from the picture archiving and communication system (PACS), automatically curated, and used for semisupervised training of a brain MRI tumor detection model. Materials and Methods In this retrospective study, the cancer center PACS was mined for brain MRI scans acquired between January 2012 and December 2017 and included all annotated axial T1 postcontrast images. Line annotations were converted to boxes, excluding boxes shorter than 1 cm or longer than 7 cm. The resulting boxes were used for supervised training of object detection models using RetinaNet and Mask region-based convolutional neural network (R-CNN) architectures. The best-performing model trained from the mined data set was used to detect unannotated tumors on training images themselves (self-labeling), automatically correcting many of the missing labels. After self-labeling, new models were trained using this expanded data set. Models were scored for precision, recall, and F1 using a held-out test data set comprising 754 manually labeled images from 100 patients (403 intra-axial and 56 extra-axial enhancing tumors). Model F1 scores were compared using bootstrap resampling. Results The PACS query extracted 31 150 line annotations, yielding 11 880 boxes that met inclusion criteria. This mined data set was used to train models, yielding F1 scores of 0.886 for RetinaNet and 0.908 for Mask R-CNN. Self-labeling added 18 562 training boxes, improving model F1 scores to 0.935 (P < .001) and 0.954 (P < .001), respectively. Conclusion The application of semisupervised learning to mined image annotations significantly improved tumor detection performance, achieving an excellent F1 score of 0.954. This development pipeline can be extended for other imaging modalities, repurposing unused data silos to potentially enable automated tumor detection across radiologic modalities. © RSNA, 2022 Online supplemental material is available for this article.
Asunto(s)
Inteligencia Artificial , Redes Neurales de la Computación , Encéfalo , Humanos , Imagen por Resonancia Magnética , Estudios RetrospectivosRESUMEN
Integration of artificial intelligence (AI) applications within clinical workflows is an important step for leveraging developed AI algorithms. In this report, generalizable components for deploying AI systems into clinical practice are described that were implemented in a clinical pilot study using lymphoscintigraphy examinations as a prospective use case (July 1, 2019-October 31, 2020). Deployment of the AI algorithm consisted of seven software components, as follows: (a) image delivery, (b) quality control, (c) a results database, (d) results processing, (e) results presentation and delivery, (f) error correction, and (g) a dashboard for performance monitoring. A total of 14 users used the system (faculty radiologists and trainees) to assess the degree of satisfaction with the components and overall workflow. Analyses included the assessment of the number of examinations processed, error rates, and corrections. The AI system processed 1748 lymphoscintigraphy examinations. The system enabled radiologists to correct 146 AI results, generating real-time corrections to the radiology report. All AI results and corrections were successfully stored in a database for downstream use by the various integration components. A dashboard allowed monitoring of the AI system performance in real time. All 14 survey respondents "somewhat agreed" or "strongly agreed" that the AI system was well integrated into the clinical workflow. In all, a framework of processes and components for integrating AI algorithms into clinical workflows was developed. The implementation described could be helpful for assessing and monitoring AI performance in clinical practice. Keywords: PACS, Computer Applications-General (Informatics), Diagnosis © RSNA, 2021.
RESUMEN
Federated learning (FL) is a method used for training artificial intelligence models with data from multiple sources while maintaining data anonymity, thus removing many barriers to data sharing. Here we used data from 20 institutes across the globe to train a FL model, called EXAM (electronic medical record (EMR) chest X-ray AI model), that predicts the future oxygen requirements of symptomatic patients with COVID-19 using inputs of vital signs, laboratory data and chest X-rays. EXAM achieved an average area under the curve (AUC) >0.92 for predicting outcomes at 24 and 72 h from the time of initial presentation to the emergency room, and it provided 16% improvement in average AUC measured across all participating sites and an average increase in generalizability of 38% when compared with models trained at a single site using that site's data. For prediction of mechanical ventilation treatment or death at 24 h at the largest independent test site, EXAM achieved a sensitivity of 0.950 and specificity of 0.882. In this study, FL facilitated rapid data science collaboration without data exchange and generated a model that generalized across heterogeneous, unharmonized datasets for prediction of clinical outcomes in patients with COVID-19, setting the stage for the broader use of FL in healthcare.
Asunto(s)
COVID-19/fisiopatología , Aprendizaje Automático , Evaluación de Resultado en la Atención de Salud , COVID-19/terapia , COVID-19/virología , Registros Electrónicos de Salud , Humanos , Pronóstico , SARS-CoV-2/aislamiento & purificaciónRESUMEN
PURPOSE: Automatic localization of pneumonia on chest X-rays (CXRs) is highly desirable both as an interpretive aid to the radiologist and for timely diagnosis of the disease. However, pneumonia's amorphous appearance on CXRs and complexity of normal anatomy in the chest present key challenges that hinder accurate localization. Existing studies in this area are either not optimized to preserve spatial information of abnormality or depend on expensive expert-annotated bounding boxes. We present a novel generative adversarial network (GAN)-based machine learning approach for this problem, which is weakly supervised (does not require any location annotations), was trained to retain spatial information, and can produce pixel-wise abnormality maps highlighting regions of abnormality (as opposed to bounding boxes around abnormality). METHODS: Our method is based on the Wasserstein GAN framework and, to the best of our knowledge, the first application of GANs to this problem. Specifically, from an abnormal CXR as input, we generated the corresponding pseudo normal CXR image as output. The pseudo normal CXR is the "hypothetical" normal, if the same abnormal CXR were not to have any abnormalities. We surmise that the difference between the pseudo normal and the abnormal CXR highlights the pixels suspected to have pneumonia and hence is our output abnormality map. We trained our algorithm on an "unpaired" data set of abnormal and normal CXRs and did not require any location annotations such as bounding boxes/segmentations of abnormal regions. Furthermore, we incorporated additional prior knowledge/constraints into the model and showed that they help improve localization performance. We validated the model on a data set consisting of 14 184 CXRs from the Radiological Society of North America pneumonia detection challenge. RESULTS: We evaluated our methods by comparing the generated abnormality maps with radiologist annotated bounding boxes using receiver operating characteristic (ROC) analysis, image similarity metrics such as normalized cross-correlation/mutual information, and abnormality detection rate.We also present visual examples of the abnormality maps, covering various scenarios of abnormality occurrence. Results demonstrate the ability to highlight regions of abnormality with the best method achieving an ROC area under the curve (AUC) of 0.77 and a detection rate of 85%.The GAN tended to perform better as prior knowledge/constraints were incorporated into the model. CONCLUSIONS: We presented a novel GAN based approach for localizing pneumonia on CXRs that (1) does not require expensive hand annotated location ground truth; and (2) was trained to produce abnormality maps at the pixel level as opposed to bounding boxes. We demonstrated the efficacy of our methods via quantitative and qualitative results.
Asunto(s)
Neumonía , Algoritmos , Humanos , Neumonía/diagnóstico por imagen , Curva ROC , Radiografía , Rayos XRESUMEN
Natural language processing (NLP) is a methodology designed to extract concepts and meaning from human-generated unstructured (free-form) text. It is intended to be implemented by using computer algorithms so that it can be run on a corpus of documents quickly and reliably. To enable machine learning (ML) techniques in NLP, free-form text must be converted to a numerical representation. After several stages of preprocessing including tokenization, removal of stop words, token normalization, and creation of a master dictionary, the bag-of-words (BOW) technique can be used to represent each remaining word as a feature of the document. The preprocessing steps simplify the documents but also potentially degrade meaning. The values of the features in BOW can be modified by using techniques such as term count, term frequency, and term frequency-inverse document frequency. Experience and experimentation will guide decisions on which specific techniques will optimize ML performance. These and other NLP techniques are being applied in radiology. Radiologists' understanding of the strengths and limitations of these techniques will help in communication with data scientists and in implementation for specific tasks. Online supplemental material is available for this article. ©RSNA, 2021.
Asunto(s)
Procesamiento de Lenguaje Natural , Radiología , Algoritmos , Humanos , Aprendizaje Automático , RadiólogosRESUMEN
Background Patterns of metastasis in cancer are increasingly relevant to prognostication and treatment planning but have historically been documented by means of autopsy series. Purpose To show the feasibility of using natural language processing (NLP) to gather accurate data from radiology reports for assessing spatial and temporal patterns of metastatic spread in a large patient cohort. Materials and Methods In this retrospective longitudinal study, consecutive patients who underwent CT from July 2009 to April 2019 and whose CT reports followed a departmental structured template were included. Three radiologists manually curated a sample of 2219 reports for the presence or absence of metastases across 13 organs; these manually curated reports were used to develop three NLP models with an 80%-20% split for training and test sets. A separate random sample of 448 manually curated reports was used for validation. Model performance was measured by accuracy, precision, and recall for each organ. The best-performing NLP model was used to generate a final database of metastatic disease across all patients. For each cancer type, statistical descriptive reports were provided by analyzing the frequencies of metastatic disease at the report and patient levels. Results In 91 665 patients (mean age ± standard deviation, 61 years ± 15; 46 939 women), 387 359 reports were labeled. The best-performing NLP model achieved accuracies from 90% to 99% across all organs. Metastases were most frequently reported in abdominopelvic (23.6% of all reports) and thoracic (17.6%) nodes, followed by lungs (14.7%), liver (13.7%), and bones (9.9%). Metastatic disease tropism is distinct among common cancers, with the most common first site being bones in prostate and breast cancers and liver among pancreatic and colorectal cancers. Conclusion Natural language processing may be applied to cancer patients' CT reports to generate a large database of metastatic phenotypes. Such a database could be combined with genomic studies and used to explore prognostic imaging phenotypes with relevance to treatment planning. © RSNA, 2021 Online supplemental material is available for this article.
Asunto(s)
Manejo de Datos/métodos , Bases de Datos Factuales/estadística & datos numéricos , Registros Electrónicos de Salud , Procesamiento de Lenguaje Natural , Neoplasias/epidemiología , Tomografía Computarizada por Rayos X/métodos , Estudios de Factibilidad , Femenino , Humanos , Estudios Longitudinales , Masculino , Persona de Mediana Edad , Metástasis de la Neoplasia , Reproducibilidad de los Resultados , Estudios RetrospectivosRESUMEN
PURPOSE: Immune checkpoint inhibition (ICI) alone is not active in mismatch repair-proficient (MMR-P) metastatic colorectal cancer (mCRC), nor does radiotherapy alone result in objective systemic benefit. However, combined radiotherapy plus ICI can induce systemic antitumor immunity in preclinical and clinical models. PATIENTS AND METHODS: In this single-center, phase II study, patients with chemotherapy-refractory MMR-P mCRC received durvalumab 1,500 mg plus tremelimumab 75 mg every 4 weeks plus radiotherapy. The primary endpoint was objective response rate (ORR) in nonirradiated lesions. Treatment and efficacy were correlated with peripheral immune cell profiles. RESULTS: We enrolled 24 patients, and report outcomes after a median follow-up of 21.8 (range: 15.9-26.3) months. The ORR was 8.3% (2 patients) [95% confidence interval (CI), 1.0-27.0]. The median progression-free survival was 1.8 (95% CI, 1.7-1.9) months, median overall survival was 11.4 (95% CI, 10.1-17.4) months. Twenty five percent of patients (n = 6) had treatment-related grade 3-4 adverse events. We observed increased circulating CD8+ T lymphocyte activation, differentiation, and proliferation in patients with objective response. CONCLUSIONS: This combination of radiotherapy plus ICI study did not meet the prespecified endpoint criteria to be considered worthwhile for further study. However, rare instances of systemic immune augmentation and regression in nonirradiated lesions were observed (an abscopal response). Combination durvalumab and tremelimumab plus radiotherapy is feasible in MMR-P mCRC with a manageable safety profile. Further studies of novel immunotherapy combinations, and identification of biomarkers predictive of abscopal response are warranted.
Asunto(s)
Protocolos de Quimioterapia Combinada Antineoplásica/administración & dosificación , Quimioradioterapia/métodos , Neoplasias Colorrectales/terapia , Inhibidores de Puntos de Control Inmunológico/administración & dosificación , Adulto , Anciano , Anticuerpos Monoclonales/administración & dosificación , Anticuerpos Monoclonales/efectos adversos , Anticuerpos Monoclonales Humanizados/administración & dosificación , Anticuerpos Monoclonales Humanizados/efectos adversos , Protocolos de Quimioterapia Combinada Antineoplásica/efectos adversos , Linfocitos T CD8-positivos/inmunología , Quimioradioterapia/efectos adversos , Neoplasias Colorrectales/inmunología , Neoplasias Colorrectales/mortalidad , Neoplasias Colorrectales/patología , Reparación de la Incompatibilidad de ADN/inmunología , Estudios de Factibilidad , Femenino , Estudios de Seguimiento , Humanos , Inhibidores de Puntos de Control Inmunológico/efectos adversos , Masculino , Persona de Mediana Edad , Supervivencia sin Progresión , Criterios de Evaluación de Respuesta en Tumores SólidosRESUMEN
'Federated Learning' (FL) is a method to train Artificial Intelligence (AI) models with data from multiple sources while maintaining anonymity of the data thus removing many barriers to data sharing. During the SARS-COV-2 pandemic, 20 institutes collaborated on a healthcare FL study to predict future oxygen requirements of infected patients using inputs of vital signs, laboratory data, and chest x-rays, constituting the "EXAM" (EMR CXR AI Model) model. EXAM achieved an average Area Under the Curve (AUC) of over 0.92, an average improvement of 16%, and a 38% increase in generalisability over local models. The FL paradigm was successfully applied to facilitate a rapid data science collaboration without data exchange, resulting in a model that generalised across heterogeneous, unharmonized datasets. This provided the broader healthcare community with a validated model to respond to COVID-19 challenges, as well as set the stage for broader use of FL in healthcare.
RESUMEN
Venous thromboembolism (VTE) associated with cancer (CAT) is a well-described complication of cancer and a leading cause of death in patients with cancer. The purpose of this study was to assess potential associations of molecular signatures with CAT, including tumor-specific mutations and the presence of clonal hematopoiesis. We analyzed deep-coverage targeted DNA-sequencing data of >14 000 solid tumor samples using the Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets platform to identify somatic alterations associated with VTE. End point was defined as the first instance of cancer-associated pulmonary embolism and/or proximal/distal lower extremity deep vein thrombosis. Cause-specific Cox proportional hazards regression was used, adjusting for pertinent clinical covariates. Of 11 695 evaluable individuals, 72% had metastatic disease at time of analysis. Tumor-specific mutations in KRAS (hazard ratio [HR], 1.34; 95% confidence interval (CI), 1.09-1.64; adjusted P = .08), STK11 (HR, 2.12; 95% CI, 1.55-2.89; adjusted P < .001), KEAP1 (HR, 1.84; 95% CI, 1.21-2.79; adjusted P = .07), CTNNB1 (HR, 1.73; 95% CI, 1.15-2.60; adjusted P = .09), CDKN2B (HR, 1.45; 95% CI, 1.13-1.85; adjusted P = .07), and MET (HR, 1.83; 95% CI, 1.15-2.92; adjusted P = .09) were associated with a significantly increased risk of CAT independent of tumor type. Mutations in SETD2 were associated with a decreased risk of CAT (HR, 0.35; 95% CI, 0.16-0.79; adjusted P = .09). The presence of clonal hematopoiesis was not associated with an increased VTE rate. This is the first large-scale analysis to elucidate tumor-specific genomic events associated with CAT. Somatic tumor mutations of STK11, KRAS, CTNNB1, KEAP1, CDKN2B, and MET were associated with an increased risk of VTE in patients with solid tumors. Further analysis is needed to validate these findings and identify additional molecular signatures unique to individual tumor types.
Asunto(s)
Neoplasias/complicaciones , Tromboembolia Venosa/etiología , Anciano , Predisposición Genética a la Enfermedad , Genómica , Humanos , Persona de Mediana Edad , Mutación , Neoplasias/genética , Factores de Riesgo , Tromboembolia Venosa/genéticaRESUMEN
BACKGROUND: The intestinal microbiota contributes to the pathogenesis of obesity and metabolic disorders. People living with HIV (PLWH) have a higher risk for the development of visceral adiposity with accompanying worsened cardiovascular risk. SETTING: Convenience sample from an HIV clinic and research unit. METHODS: To understand the relationship between adiposity and intestinal dysbiosis, we compared the gut microbiota and inflammatory markers in a cross-sectional study of viscerally obese, generally obese, and lean PLWH. Fecal intestinal microbiota was characterized by 16S ribosomal DNA sequencing. Abdominal CTs quantified subcutaneous adipose tissue and visceral adipose tissue (SAT; VAT). Serum high sensitivity C-reactive protein, adiponectin, leptin, IL-6, MCP-1, and sCD14 were assayed. RESULTS: We studied 15, 9, and 11 participants with visceral obesity, general obesity, and lean body type, respectively. The generally obese group were all women and 2/3 African American, whereas the visceral obesity and lean groups were predominantly white and men who have sex with men. Markers of systemic inflammation and sCD14 were higher in general obesity compared with lean. sCD14 was positively correlated with VAT, but not SAT. Bacterial diversity was significantly reduced in participants with visceral and general obesity and composition of intestinal microbiota was significantly different from lean body types. Bacterial alpha diversity was negatively correlated with VAT area, waist/hip ratio, and sCD14, but not with SAT area. CONCLUSIONS: In this exploratory study, obesity in general was associated with dysbiotic intestinal microbiota. The relationships of VAT to bacterial diversity and sCD14 suggest that dysbiosis in viscerally obese PLWH could be associated with heightened inflammatory state.