RESUMO
OBJECTIVES: To evaluate the performance of machine learning models in predicting treatment response to neoadjuvant chemoradiotherapy (nCRT) in rectal cancer using computed tomography (CT) and magnetic resonance imaging (MRI). METHODS: We searched PubMed, Embase, Cochrane Library, and Web of Science for studies published before January 2023. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) was used to assess the methodological quality of the included studies, random-effects models were used to calculate sensitivity and specificity, I2 values were used for heterogeneity measurements, and subgroup analyses were carried out to detect potential sources of heterogeneity. RESULTS: A total of 1690 patients from 24 studies were included. The meta-analysis calculated a pooled area under the curve (AUC) of 0.92 (95%CI-0.89-0.94), pooled sensitivity of 0.81 (95%CI-0.73-0.88), and pooled specificity of 0.88 (95%CI-0.82-0.92). We investigated 4 studies that mainly contributed to heterogeneity. After performing meta-analysis again excluding these 4 studies, the heterogeneity was significantly reduced. In subgroup analysis, the pooled AUC of the deep learning model was 0.95 and was 0.88 for the traditional statistical model; the pooled AUC of studies that used diffusion-weighted imaging (DWI) was 0.90, and was 0.92 in studies that did not use DWI; the pooled AUC of studies conducted in China was 0.94, and was 0.83 in studies conducted in other countries. CONCLUSIONS: Machine learning has promising potential in predicting tumor response to nCRT in patients with locally advanced rectal cancer. Together with clinical information, machine-learning based models may bring us closer toward precision medicine. ADVANCES IN KNOWLEDGE: Compared to traditional machine learning models, deep learning-based studies are able to obtain higher AUC, although they are less predominant and more heterogeneous. Together with clinical information, machine-learning based models may bring us closer toward precision medicine.
RESUMO
Total intracranial volume (TIV) is the volume enclosed inside the cranium, inclusive of the meninges and the brain. TIV is extensively used to correct variations in inter-subject head size for the evaluation of neurodegenerative diseases. In this work, we present an automatic method to generate a TIV mask from MR images while synthesizing a CT image to be used in subsequent analysis. In addition, we propose an alternative way to obtain ground truth TIV masks using a semi-manual approach, which results in significant time savings. We train a conditional generative adversarial network (cGAN) using 2D MR slices to realize our tasks. The quantitative evaluation showed that the model was able to synthesize CT and generate TIV masks that closely approximate the reference images. This study also provides a comparison of the described method against skull stripping tools that output a mask enclosing the cranial volume, using MRI scan. In particular, highlighting the deficiencies in using such tools to approximate the volume using MRI scan.
RESUMO
An IEEE Engineering in Medicine and Biology Society (EMBS) student chapter can play an important service role for collegiate biomedical curricula, supporting (a) faculty and administrators as they offer biomedical programs and work to strengthen industry/community relationships, and (b) students as they engage in engineering skill development and seek industry employment, graduate school opportunities, or medical school placement. This paper summarizes recent projects and activities sponsored by the Kansas State University (KSU) Student Chapter of the IEEE EMBS - efforts intended to maintain interest in the student chapter while supporting its service role. Such a role will become more important in upcoming years in light of the increasing demand for biomedical engineers, especially in the Midwest United States, a reality which motivated the inception of a new KSU undergraduate degree in Biomedical Engineering starting in Fall 2018. The KSU IEEE EMBS student chapter can play a large role in the overall success of this new curriculum, and the projects and activities summarized in this paper are offered as examples to programs that may wish to benefit from an IEEE EMBS student chapter in a similar and meaningful way.
Assuntos
Engenharia Biomédica , Sociedades/organização & administração , Estudantes , Currículo , Humanos , Estados Unidos , UniversidadesRESUMO
Microwave plasma torch (MPT) is a simple and low power-consumption ambient ion source. And the MPT Mass spectra of many metal elements usually exhibit some novel features different from their inductively coupled plasma (ICP) mass spectra, which may be helpful for metal element analysis. Here, we presented the results about the MPT mass spectra of copper and molybdenum elements by a linear ion trap mass spectrometer (LTQ). The generated copper or molybdenum contained ions in plasma were characterized further in collision-induced dissociated (CID) experiments. These researches built a novel, direct and sensitive method for the direct analysis of trace levels of copper and molybdenum in aqueous liquids. Quantitative results showed that the limit of detection (LOD) by using MS(2) procedure was estimated to be 0.265 µg/l (ppb) for copper and 0.497 µg/l for molybdenum. The linear dynamics ranges cover at least 2 orders of magnitude and the analysis of a single aqueous sample can be completed in 5-6 min with a reasonable semi-quantitative sense. Two practical aqueous samples, milk and urine, were also analyzed qualitatively with reasonable recovery rates and RSD. These experimental data demonstrated that the MPT MS is able to turn into a promising and hopeful tool in field analysis of copper and molybdenum ions in water and some aqueous media, and can be applied in many fields, such as environmental controlling, hydrogeology, and water quality inspection. Moreover, MPT MS could also be used as the supplement of ICP-MS for the rapid and in-situ analysis of metal ions. Copyright © 2016 John Wiley & Sons, Ltd.
