scholarly journals On Training Deep 3D CNN Models with Dependent Samples in Neuroimaging

Author(s):  
Yunyang Xiong ◽  
Hyunwoo J. Kim ◽  
Bhargav Tangirala ◽  
Ronak Mehta ◽  
Sterling C. Johnson ◽  
...  
Keyword(s):  
3D Cnn ◽  
2021 ◽  
Vol 9 (1) ◽  
pp. 20-31
Author(s):  
Hasan Huseyin Aksu ◽  

The purpose of this study is to answer if there is a reasonable difference on academical success of students who get education with traditional and RME approach question on “Teaching geometrical objects to 8th grade students” subject. Study group consists of 47 students which contains 21 experimental and 16 control group from “Ordu Anadolu İmam Hatip High School Project School” in Altinordu, Ordu. Experimental and control group have same academical success level, as the school which this study has runned is a school which accepts students with an exam only. After the experimental and control groups were created, a 25 question pre-test was performed to understand the level of knowledge of the group regarding geometrical objects. The same test was performed on the same groups 8 weeks later as retention test. To determine opinions of the students in experimental group regarding RME and related learning activities, semi-structured interviews are conducted. The data obtained from the pretest, posttest and retention tests were analyzed with t-test for independent samples and t-test for dependent samples and variance analysis for mixed measurements with 0.05 significance level. According to the results, it is seen that learning activities prepared according to RME approach are much more effective than learning activities prepared according to the traditional approach on students’ academic success.


2018 ◽  
Vol 23 (suppl_1) ◽  
pp. e16-e16
Author(s):  
Ahmed Moussa ◽  
Audrey Larone-Juneau ◽  
Laura Fazilleau ◽  
Marie-Eve Rochon ◽  
Justine Giroux ◽  
...  

Abstract BACKGROUND Transitions to new healthcare environments can negatively impact patient care and threaten patient safety. Immersive in situ simulation conducted in newly constructed single family room (SFR) Neonatal Intensive Care Units (NICUs) prior to occupancy, has been shown to be effective in testing new environments and identifying latent safety threats (LSTs). These simulations overlay human factors to identify LSTs as new and existing process and systems are implemented in the new environment OBJECTIVES We aimed to demonstrate that large-scale, immersive, in situ simulation prior to the transition to a new SFR NICU improves: 1) systems readiness, 2) staff preparedness, 3) patient safety, 4) staff comfort with simulation, and 5) staff attitude towards culture change. DESIGN/METHODS Multidisciplinary teams of neonatal healthcare providers (HCP) and parents of former NICU patients participated in large-scale, immersive in-situ simulations conducted in the new NICU prior to occupancy. One eighth of the NICU was outfitted with equipment and mannequins and staff performed in their native roles. Multidisciplinary debriefings, which included parents, were conducted immediately after simulations to identify LSTs. Through an iterative process issues were resolved and additional simulations conducted. Debriefings were documented and debriefing transcripts transcribed and LSTs classified using qualitative methods. To assess systems readiness and staff preparedness for transition into the new NICU, HCPs completed surveys prior to transition, post-simulation and post-transition. Systems readiness and staff preparedness were rated on a 5-point Likert scale. Average survey responses were analyzed using dependent samples t-tests and repeated measures ANOVAs. RESULTS One hundred eight HCPs and 24 parents participated in six half-day simulation sessions. A total of 75 LSTs were identified and were categorized into eight themes: 1) work organization, 2) orientation and parent wayfinding, 3) communication devices/systems, 4) nursing and resuscitation equipment, 5) ergonomics, 6) parent comfort; 7) work processes, and 8) interdepartmental interactions. Prior to the transition to the new NICU, 76% of the LSTs were resolved. Survey response rate was 31%, 16%, 7% for baseline, post-simulation and post-move surveys, respectively. System readiness at baseline was 1.3/5,. Post-simulation systems readiness was 3.5/5 (p = 0.0001) and post-transition was 3.9/5 (p = 0.02). Staff preparedness at baseline was 1.4/5. Staff preparedness post-simulation was 3.3/5 (p = 0.006) and post-transition was 3.9/5 (p = 0.03). CONCLUSION Large-scale, immersive in situ simulation is a feasible and effective methodology for identifying LSTs, improving systems readiness and staff preparedness in a new SFR NICU prior to occupancy. However, to optimize patient safety, identified LSTs must be mitigated prior to occupancy. Coordinating large-scale simulations is worth the time and cost investment necessary to optimize systems and ensure patient safety prior to transition to a new SFR NICU.


Author(s):  
Alexander Bigalke ◽  
Lasse Hansen ◽  
Jasper Diesel ◽  
Mattias P. Heinrich

Abstract Purpose Body weight is a crucial parameter for patient-specific treatments, particularly in the context of proper drug dosage. Contactless weight estimation from visual sensor data constitutes a promising approach to overcome challenges arising in emergency situations. Machine learning-based methods have recently been shown to perform accurate weight estimation from point cloud data. The proposed methods, however, are designed for controlled conditions in terms of visibility and position of the patient, which limits their practical applicability. In this work, we aim to decouple accurate weight estimation from such specific conditions by predicting the weight of covered patients from voxelized point cloud data. Methods We propose a novel deep learning framework, which comprises two 3D CNN modules solving the given task in two separate steps. First, we train a 3D U-Net to virtually uncover the patient, i.e. to predict the patient’s volumetric surface without a cover. Second, the patient’s weight is predicted from this 3D volume by means of a 3D CNN architecture, which we optimized for weight regression. Results We evaluate our approach on a lying pose dataset (SLP) under two different cover conditions. The proposed framework considerably improves on the baseline model by up to $${16}{\%}$$ 16 % and reduces the gap between the accuracy of weight estimates for covered and uncovered patients by up to $${52}{\%}$$ 52 % . Conclusion We present a novel pipeline to estimate the weight of patients, which are covered by a blanket. Our approach relaxes the specific conditions that were required for accurate weight estimates by previous contactless methods and thus constitutes an important step towards fully automatic weight estimation in clinical practice.


Author(s):  
Bin Sun ◽  
Fengyin Liu ◽  
Yusun Zhou ◽  
Shaolei Jin ◽  
Qiang Li ◽  
...  
Keyword(s):  

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Thao Thi Ho ◽  
Taewoo Kim ◽  
Woo Jin Kim ◽  
Chang Hyun Lee ◽  
Kum Ju Chae ◽  
...  

AbstractChronic obstructive pulmonary disease (COPD) is a respiratory disorder involving abnormalities of lung parenchymal morphology with different severities. COPD is assessed by pulmonary-function tests and computed tomography-based approaches. We introduce a new classification method for COPD grouping based on deep learning and a parametric-response mapping (PRM) method. We extracted parenchymal functional variables of functional small airway disease percentage (fSAD%) and emphysema percentage (Emph%) with an image registration technique, being provided as input parameters of 3D convolutional neural network (CNN). The integrated 3D-CNN and PRM (3D-cPRM) achieved a classification accuracy of 89.3% and a sensitivity of 88.3% in five-fold cross-validation. The prediction accuracy of the proposed 3D-cPRM exceeded those of the 2D model and traditional 3D CNNs with the same neural network, and was comparable to that of 2D pretrained PRM models. We then applied a gradient-weighted class activation mapping (Grad-CAM) that highlights the key features in the CNN learning process. Most of the class-discriminative regions appeared in the upper and middle lobes of the lung, consistent with the regions of elevated fSAD% and Emph% in COPD subjects. The 3D-cPRM successfully represented the parenchymal abnormalities in COPD and matched the CT-based diagnosis of COPD.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ossama Mahmoud ◽  
Mahmoud El-Sakka ◽  
Barry G. H. Janssen

AbstractMicrovascular blood flow is crucial for tissue and organ function and is often severely affected by diseases. Therefore, investigating the microvasculature under different pathological circumstances is essential to understand the role of the microcirculation in health and sickness. Microvascular blood flow is generally investigated with Intravital Video Microscopy (IVM), and the captured images are stored on a computer for later off-line analysis. The analysis of these images is a manual and challenging process, evaluating experiments very time consuming and susceptible to human error. Since more advanced digital cameras are used in IVM, the experimental data volume will also increase significantly. This study presents a new two-step image processing algorithm that uses a trained Convolutional Neural Network (CNN) to functionally analyze IVM microscopic images without the need for manual analysis. While the first step uses a modified vessel segmentation algorithm to extract the location of vessel-like structures, the second step uses a 3D-CNN to assess whether the vessel-like structures have blood flowing in it or not. We demonstrate that our two-step algorithm can efficiently analyze IVM image data with high accuracy (83%). To our knowledge, this is the first application of machine learning for the functional analysis of microvascular blood flow in vivo.


Author(s):  
Chengyang Li ◽  
Liping Zhu ◽  
Dandan Zhu ◽  
Jiale Chen ◽  
Zhanghui Pan ◽  
...  
Keyword(s):  

2015 ◽  
Vol 2015 ◽  
pp. 1-8
Author(s):  
Mingchen Yao ◽  
Chao Zhang ◽  
Wei Wu

Many generalization results in learning theory are established under the assumption that samples are independent and identically distributed (i.i.d.). However, numerous learning tasks in practical applications involve the time-dependent data. In this paper, we propose a theoretical framework to analyze the generalization performance of the empirical risk minimization (ERM) principle for sequences of time-dependent samples (TDS). In particular, we first present the generalization bound of ERM principle for TDS. By introducing some auxiliary quantities, we also give a further analysis of the generalization properties and the asymptotical behaviors of ERM principle for TDS.


2021 ◽  
pp. 100709
Author(s):  
Md. Kamrul Hasan ◽  
Md. Tasnim Jawad ◽  
Kazi Nasim Imtiaz Hasan ◽  
Sajal Basak Partha ◽  
Md. Masum Al Masba ◽  
...  
Keyword(s):  
Chest Ct ◽  
Ct Scans ◽  

Sign in / Sign up

Export Citation Format

Share Document