RESUMO
INTRODUCTION AND HYPOTHESIS: The objective was to create and validate the usefulness of a convolutional neural network (CNN) for identifying different organs of the pelvic floor in the midsagittal plane via dynamic ultrasound. METHODS: This observational and prospective study included 110 patients. Transperineal ultrasound scans were performed by an expert sonographer of the pelvic floor. A video of each patient was made that captured the midsagittal plane of the pelvic floor at rest and the change in the pelvic structures during the Valsalva maneuver. After saving the captured videos, we manually labeled the different organs in each video. Three different architectures were tested-UNet, FPN, and LinkNet-to determine which CNN model best recognized anatomical structures. The best model was trained with the 86 cases for the number of epochs determined by the stop criterion via cross-validation. The Dice Similarity Index (DSI) was used for CNN validation. RESULTS: Eighty-six patients were included to train the CNN and 24 to test the CNN. After applying the trained CNN to the 24 test videos, we did not observe any failed segmentation. In fact, we obtained a DSI of 0.79 (95% CI: 0.73 - 0.82) as the median of the 24 test videos. When we studied the organs independently, we observed differences in the DSI of each organ. The poorest DSIs were obtained in the bladder (0.71 [95% CI: 0.70 - 0.73]) and uterus (0.70 [95% CI: 0.68 - 0.74]), whereas the highest DSIs were obtained in the anus (0.81 [95% CI: 0.80 - 0.86]) and levator ani muscle (0.83 [95% CI: 0.82 - 0.83]). CONCLUSIONS: Our results show that it is possible to apply deep learning using a trained CNN to identify different pelvic floor organs in the midsagittal plane via dynamic ultrasound.
RESUMO
PURPOSE: There remains a need for novel therapies for patients with metastatic breast cancer (MBC). We explore the use of a novel biomarker of survival that could potentially expedite the testing of novel therapies. METHODS: We applied a tumor regression-growth model to radiographic measurement data from 393 women with MBC enrolled in PALOMA-3 examining efficacy of palbociclib in disease that had progressed on previous endocrine therapy. 261 and 132 women were randomized to fulvestrant plus palbociclib or placebo, respectively. We estimated rates of regression (d) and growth (g) of the sensitive and resistant fractions of tumors, respectively. We compared the median g of both arms. We examined the relationship between g and progression-free and overall survival (OS). RESULTS: As in other tumors, g is a biomarker of OS. In PALOMA-3, we found significant differences in g among patients with tumors sensitive to endocrine therapy but not amongst resistant tumors, emulating clinical trial results. Subgroup analysis found favorable g values in visceral metastases treated with palbociclib. Palbociclib efficacy demonstrated by slower g values was evident early in the trial, twelve weeks after the first 28 patients had been enrolled. CONCLUSION: Values of g, estimated using data collected while a patient is enrolled in a clinical trial is an excellent biomarker of OS. Our results correlate with the survival outcomes of PALOMA-3 and argue strongly for using g as a clinical trial endpoint to help inform go/no-go decisions, improve trial efficiency, and deliver novel therapies to patients sooner.
Assuntos
Neoplasias da Mama , Piridinas , Feminino , Humanos , Biomarcadores , Neoplasias da Mama/tratamento farmacológico , Neoplasias da Mama/patologia , Intervalo Livre de Doença , Piperazinas , Piridinas/uso terapêutico , Ensaios Clínicos Controlados Aleatórios como Assunto , Receptor ErbB-2RESUMO
BACKGROUND: The impact of such recommendations after their implementation of guidelines has not usually been evaluated. Herein, we assessed the impact and compliance with the Spanish Oncology Genitourinary Group (SOGUG) Guidelines for toxicity management of targeted therapies in metastatic renal cell carcinoma (mRCC) in daily clinical practice. METHODS: Data on 407 mRCC patients who initiated first-line targeted therapy during the year before and the year after publication and implementation of the SOGUG guideline program were available from 34 Spanish Hospitals. Adherence to SOGUG Guidelines was assessed in every cycle. RESULTS: Adverse event (AE) management was consistent with the Guidelines as a whole for 28.7% out of 966 post-implementation cycles compared with 23.1% out of 892 pre-implementation cycles (p = 0.006). Analysis of adherence by AE in non-compliant cycles showed significant changes in appropriate management of hypertension (33% pre-implementation vs. 44.5% post-implementation cycles; p < 0.0001), diarrhea (74.0% vs. 80.5%; p = 0.011) and dyslipemia (25.0% vs. 44.6%; p < 0.001). CONCLUSIONS: Slight but significant improvements in AE management were detected following the implementation of SOGUG recommendations. However, room for improvement in the management of AEs due to targeted agents still remains and could be the focus for further programs in this direction.
Assuntos
Antineoplásicos/efeitos adversos , Carcinoma de Células Renais/tratamento farmacológico , Neoplasias Renais/tratamento farmacológico , Idoso , Antineoplásicos/uso terapêutico , Feminino , Fidelidade a Diretrizes/estatística & dados numéricos , Humanos , Masculino , Pessoa de Meia-Idade , Terapia de Alvo Molecular/efeitos adversos , Metástase Neoplásica , Guias de Prática Clínica como Assunto , EspanhaRESUMO
It was found that the diffraction images acquired along the side scattering directions with objects in a cell sample contain pattern variations at both the global and local scales. We show here that the global pattern variation is associated with the categorical size and morphological heterogeneity of the imaged objects. An automated image processing method has been developed to separate the acquired diffraction images into three types of global patterns. Combined with previously developed method for quantifying local texture pattern variations, the new method allows fully automated analysis of diffraction images for rapid and label-free classification of cells according to their 3D morphology.
Assuntos
Fenômenos Fisiológicos Celulares , Separação Celular/métodos , Citometria de Fluxo/métodos , Aumento da Imagem/métodos , Interpretação de Imagem Assistida por Computador/métodos , Imageamento Tridimensional/métodos , Refratometria/métodos , AlgoritmosRESUMO
We report a novel method of diffraction imaging flow cytometry to measure and analyze size distribution of microspheres. An automated and robust image processing software based on the short-time-Fourier-transform algorithm has been developed to analyze the characteristic and spatially varying oscillations of side scatters recorded as a diffraction image. Our results demonstrate that the new method allows accurate and rapid determination of single microspheres' diameters ranging from 1 to 100 µm. The capacity for analysis of light scattering by two-sphere aggregates has been demonstrated but analytical tools for characterization of aggregates by multiple microspheres remain to be developed.