Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572222
J. Flessner, M. Frenken
This work describes a high level modeling approach of building automation and control systems (BACS) using perceptual knowledge. Present knowledge about the human perception is advantageous for the development of human-centered BACS. However, there exists no general approach describing the involvement of perceptual knowledge within the modeling process of BACS. In this paper, a novel approach for the development of a perceptual knowledge base is described. The development of the knowledge base uses the structure of finite state machines. The purpose of the knowledge base is the derivation of an abstract rule base which forms a useful framework for the design of BACS applications. In conclusion, the presented approach supports the design of BACS with the focus on the human perception.
{"title":"High Level Modeling of Building Automation and Control Systems Based on Perceptual Knowledge","authors":"J. Flessner, M. Frenken","doi":"10.1109/LSC.2018.8572222","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572222","url":null,"abstract":"This work describes a high level modeling approach of building automation and control systems (BACS) using perceptual knowledge. Present knowledge about the human perception is advantageous for the development of human-centered BACS. However, there exists no general approach describing the involvement of perceptual knowledge within the modeling process of BACS. In this paper, a novel approach for the development of a perceptual knowledge base is described. The development of the knowledge base uses the structure of finite state machines. The purpose of the knowledge base is the derivation of an abstract rule base which forms a useful framework for the design of BACS applications. In conclusion, the presented approach supports the design of BACS with the focus on the human perception.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"172 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124209160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572090
K. Dick, J. Green
Rank order data are pervasive in science and in our daily lived experience. With the advent of high performance computing and the commensurate increase in available data, the opportunity to capture the overall distribution of values by means of nonparametric curve fitting enables the identification of exceptional points in large datasets. With a rank order structure, these distributions may exhibit a “knee” delineating a threshold between exceptional points and those of the baseline. Given an accurate characterization of the distribution of prediction scores, including careful identification of the knee, we have previously shown that predictive performance can be significantly improved by leveraging this “context”. This paper examines the nonparametric characterization of such distributions. Locally weighted regression (LOESS) is a widely used nonparametric approach to curve fitting. Here, we revisit the assumptions behind the selection of kernel functions for nonparametric curve fitting of biological and biomedical data exhibiting rare or exceptional instances. We propose a new linear asymmetric kernel function and compare it to the commonly used tricube kernel used in LOESS. We evaluate its ability to fit rank order data in the domain of protein-protein interaction prediction. The proposed linear kernel significantly improved predictive performance $(p < 0.001$) of two state-of-the-art predictors and promises to be widely applicable in related machine learning pipelines and nonparametric regression tasks.
{"title":"Fitting Rank Order Data in the Age of Context","authors":"K. Dick, J. Green","doi":"10.1109/LSC.2018.8572090","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572090","url":null,"abstract":"Rank order data are pervasive in science and in our daily lived experience. With the advent of high performance computing and the commensurate increase in available data, the opportunity to capture the overall distribution of values by means of nonparametric curve fitting enables the identification of exceptional points in large datasets. With a rank order structure, these distributions may exhibit a “knee” delineating a threshold between exceptional points and those of the baseline. Given an accurate characterization of the distribution of prediction scores, including careful identification of the knee, we have previously shown that predictive performance can be significantly improved by leveraging this “context”. This paper examines the nonparametric characterization of such distributions. Locally weighted regression (LOESS) is a widely used nonparametric approach to curve fitting. Here, we revisit the assumptions behind the selection of kernel functions for nonparametric curve fitting of biological and biomedical data exhibiting rare or exceptional instances. We propose a new linear asymmetric kernel function and compare it to the commonly used tricube kernel used in LOESS. We evaluate its ability to fit rank order data in the domain of protein-protein interaction prediction. The proposed linear kernel significantly improved predictive performance $(p < 0.001$) of two state-of-the-art predictors and promises to be widely applicable in related machine learning pipelines and nonparametric regression tasks.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134035347","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572240
Yongjian Yu, Jue Wang
We present a system of techniques for automatic segmentation, quantification, and morphotype classification of vaginal bacteria from multi-band fluorescent microscopic imagery. Individual bacteria segmentation is accomplished via data pre-processing, blobness enhancement, thresholding, and multi-scale morphological decomposition. A new spotness feature is devised and extracted to effectively quantify bacterial morphotypes. A supervised classifier is trained on microscopic scans containing thousands of bacteria. Our approach is able to predict and segment bacteria with a high accuracy. The average classification error in terms of bacteria composition ratio is 6% relative to the ground-truth.
{"title":"Automated Enumeration and Classification of Bacteria in Fluorescent Microscopy Imagery","authors":"Yongjian Yu, Jue Wang","doi":"10.1109/LSC.2018.8572240","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572240","url":null,"abstract":"We present a system of techniques for automatic segmentation, quantification, and morphotype classification of vaginal bacteria from multi-band fluorescent microscopic imagery. Individual bacteria segmentation is accomplished via data pre-processing, blobness enhancement, thresholding, and multi-scale morphological decomposition. A new spotness feature is devised and extracted to effectively quantify bacterial morphotypes. A supervised classifier is trained on microscopic scans containing thousands of bacteria. Our approach is able to predict and segment bacteria with a high accuracy. The average classification error in terms of bacteria composition ratio is 6% relative to the ground-truth.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114539863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572187
Nicholas Fritz, Daniel W. Gulick, Jennifer Blain Christen
Stimulation has been a key technique for studying underlying mechanisms of the nervous system. Electrical stimulation has been the predominant method for eliciting desired muscle responses for decades, yet methodologies remain invasive and low in selectivity of tissue stimulated. Current injection affects all local tissue types and can lead to damaging immune responses that threaten both nerves and equipment alike. Optogenetics provides a solution for such stimulation difficulties by increasing specificity and decreasing risk to tissue. Via genetic modifications, opsins (light-sensitive proteins) are added to neurons, and can be activated by light to cause neuron excitation. Through preliminary in vivo testing in transgenic mice expressing channelrhodopsin (ChR2) we validate that multiple beams of light have an additive effect and increase the response from muscles innervated by the target nerve. Measuring hindlimb flexion increases with increase in number of light sources present. To further characterize this additive effect, a Monte Carlo computer model was generated to simulate a random-walk of photons passing through nerve tissue. The model shows that light beams can aggregate within the nerve, although are limited. When using collimated light, multiple beams converging on the interior region of the nerve cannot result in a higher intensity than outermost layer of tissue nearest a single light source. This model serves as a tool to aid future animal studies by determining light emission parameters, specifically prescribing the need for optically-focused light, when attempting to selectively stimulate regions deep in the interior of a given nerve. Such capability will allow for high spatial resolution of stimulation in peripheral nerves giving finer control of excitation in downstream tissue.
{"title":"Computational Model of Optogenetic Stimulation in a Peripheral Nerve","authors":"Nicholas Fritz, Daniel W. Gulick, Jennifer Blain Christen","doi":"10.1109/LSC.2018.8572187","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572187","url":null,"abstract":"Stimulation has been a key technique for studying underlying mechanisms of the nervous system. Electrical stimulation has been the predominant method for eliciting desired muscle responses for decades, yet methodologies remain invasive and low in selectivity of tissue stimulated. Current injection affects all local tissue types and can lead to damaging immune responses that threaten both nerves and equipment alike. Optogenetics provides a solution for such stimulation difficulties by increasing specificity and decreasing risk to tissue. Via genetic modifications, opsins (light-sensitive proteins) are added to neurons, and can be activated by light to cause neuron excitation. Through preliminary in vivo testing in transgenic mice expressing channelrhodopsin (ChR2) we validate that multiple beams of light have an additive effect and increase the response from muscles innervated by the target nerve. Measuring hindlimb flexion increases with increase in number of light sources present. To further characterize this additive effect, a Monte Carlo computer model was generated to simulate a random-walk of photons passing through nerve tissue. The model shows that light beams can aggregate within the nerve, although are limited. When using collimated light, multiple beams converging on the interior region of the nerve cannot result in a higher intensity than outermost layer of tissue nearest a single light source. This model serves as a tool to aid future animal studies by determining light emission parameters, specifically prescribing the need for optically-focused light, when attempting to selectively stimulate regions deep in the interior of a given nerve. Such capability will allow for high spatial resolution of stimulation in peripheral nerves giving finer control of excitation in downstream tissue.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121805157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572043
R. Nakamura, Y. Mitsukura
Recently many people are suffering from mental illnesses like depression worldwide. Although they are ambiguous and have difficulties in grasping the states of patients, in fact they lower their quality of life. The total loss of economics, life and quality of life by depression is big enough that it cannot be ignored. It is important for the patients to recover from depression and also for the healthy controls not to become depression. So correct diagnosis and treatment are essential for the people. In actual clinical field, incorrectness of diagnosis is now regarded as issue. To construct an objective way of evaluation on depression, we set a goal of extraction of features in depressive electroencephalography (EEG). Unlike other studies in this field, this study has mainly two points of unique. Firstly, this feature analysis is using signal from just one channel located in frontal lobe (Fp1). Secondly, the acquisition of EEG was conducted during actual clinical inquiry or under similar situation. After the experiment, EEG of both depression patients and healthy controls were compared through two-sample t-test.
{"title":"Feature Analysis of Electroencephalography in Patients with Depression","authors":"R. Nakamura, Y. Mitsukura","doi":"10.1109/LSC.2018.8572043","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572043","url":null,"abstract":"Recently many people are suffering from mental illnesses like depression worldwide. Although they are ambiguous and have difficulties in grasping the states of patients, in fact they lower their quality of life. The total loss of economics, life and quality of life by depression is big enough that it cannot be ignored. It is important for the patients to recover from depression and also for the healthy controls not to become depression. So correct diagnosis and treatment are essential for the people. In actual clinical field, incorrectness of diagnosis is now regarded as issue. To construct an objective way of evaluation on depression, we set a goal of extraction of features in depressive electroencephalography (EEG). Unlike other studies in this field, this study has mainly two points of unique. Firstly, this feature analysis is using signal from just one channel located in frontal lobe (Fp1). Secondly, the acquisition of EEG was conducted during actual clinical inquiry or under similar situation. After the experiment, EEG of both depression patients and healthy controls were compared through two-sample t-test.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"256 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122820902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572048
H. Sadreazami, M. Bolic, S. Rajan
Fail detection problem for smart home-care systems using an ultra wideband radar is considered in this paper. The goal is to identify the occurrence of fall from the radar return signals through a supervised learning approach. To this end, a new framework is proposed based on stacked long-short-term memory (LSTM) recurrent neural network to develop a robust method for feature extraction and classification of radar data of human daily activity. It is noted that the proposed method do not require heavy preprocessing on the data or feature engineering. It is known that LSTM networks are capable of capturing dependencies in time series data. In view of this, the radar time series data are directly fed into a stacked LSTM network for automatic feature extraction. Experiments are conducted on radar data collected from different subjects, when performing fall and non-fall activities. It is shown that the proposed method can provide a classification accuracy higher than that yielded by the other existing methods.
{"title":"On the Use of Ultra Wideband Radar and Stacked LSTM-RNN for at Home Fall Detection","authors":"H. Sadreazami, M. Bolic, S. Rajan","doi":"10.1109/LSC.2018.8572048","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572048","url":null,"abstract":"Fail detection problem for smart home-care systems using an ultra wideband radar is considered in this paper. The goal is to identify the occurrence of fall from the radar return signals through a supervised learning approach. To this end, a new framework is proposed based on stacked long-short-term memory (LSTM) recurrent neural network to develop a robust method for feature extraction and classification of radar data of human daily activity. It is noted that the proposed method do not require heavy preprocessing on the data or feature engineering. It is known that LSTM networks are capable of capturing dependencies in time series data. In view of this, the radar time series data are directly fed into a stacked LSTM network for automatic feature extraction. Experiments are conducted on radar data collected from different subjects, when performing fall and non-fall activities. It is shown that the proposed method can provide a classification accuracy higher than that yielded by the other existing methods.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125573621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572073
Abdelaziz Hammouche, G. Cloutier, J. Tardif, J. Meunier
Intravascular ultrasound imaging (IVUS) is an interventional cardiology technique for assessing atherosclerosis lesions in artery. This technique generates images showing the different layers of the artery and allows quantitative measurements reflecting its condition. However due to the acquisition process these images are affected by artifacts like ring-down, guide wire and shadows generated by tissue calcification. In this paper we develop a 3D algorithm based on a helical snake (active contour) for the lumen segmentation in intravascular ultrasound images. The helix snake evolves based on the analysis of the statistical properties computed on windows inside and outside the contour until it reaches the luminal border. In addition we show the influence of the ring-down artifact for the luminal border detection by adding a pre-processing step for reducing its adverse effect. The algorithm was executed on 2190 images from two clinical IVUS sequences of femoral arteries presenting the ringdown artifact. The performance of the algorithm was evaluated with respect to expert manual plots and gave a mean Hausdorff distance of 0.31 mm with overlap of 89.50 % and 94.38 % for respectively Jaccard and Dice indexes improving the result by 0.29 mm, 8.79 % and 5.36 % compared to the result without artifact removal.
{"title":"Space Curve Approach for IVUS Image Segmentation","authors":"Abdelaziz Hammouche, G. Cloutier, J. Tardif, J. Meunier","doi":"10.1109/LSC.2018.8572073","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572073","url":null,"abstract":"Intravascular ultrasound imaging (IVUS) is an interventional cardiology technique for assessing atherosclerosis lesions in artery. This technique generates images showing the different layers of the artery and allows quantitative measurements reflecting its condition. However due to the acquisition process these images are affected by artifacts like ring-down, guide wire and shadows generated by tissue calcification. In this paper we develop a 3D algorithm based on a helical snake (active contour) for the lumen segmentation in intravascular ultrasound images. The helix snake evolves based on the analysis of the statistical properties computed on windows inside and outside the contour until it reaches the luminal border. In addition we show the influence of the ring-down artifact for the luminal border detection by adding a pre-processing step for reducing its adverse effect. The algorithm was executed on 2190 images from two clinical IVUS sequences of femoral arteries presenting the ringdown artifact. The performance of the algorithm was evaluated with respect to expert manual plots and gave a mean Hausdorff distance of 0.31 mm with overlap of 89.50 % and 94.38 % for respectively Jaccard and Dice indexes improving the result by 0.29 mm, 8.79 % and 5.36 % compared to the result without artifact removal.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"50 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132283675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572202
Takuma Ando, Takashi Watanabe
Measurement of three-dimensional lower limb joint angles are useful to evaluate changes of movements after various lower limb diseases or injuries. However, estimation of three-dimensional angles with inertial measurement units (IMUs) causes errors, especially in abduction/adduction angle and in internal/external rotation angle. This was considered to be caused by difference between the sensor coordinate system and the body coordinate system. In order to solve the problem, various calibration methods of the coordinate system have been proposed. In this paper, three calibration methods, which were selected based on practical application, were examined in estimation of three-dimensional angles of a rigid body model under the 8 attachment conditions of IMU. The Method A determined the body coordinate system by measuring the vertical axis during a standing upright posture and another posture in the sagittal plane. The Method B estimated the sagittal plane and the normal vector of the plane from measured acceleration signals during movement of each segment of the lower limb in the sagittal plane. Method C was similar to Method A, in which postures of vertical and horizontal positions of lower limbs were used. Difference of the coordinate system of the IMU increased significantly RMSE values of estimated angles. Since the Method A and C showed almost same RMSE values as in the case that there was no difference of the coordinate system, the methods are considered to be practical. However, the Method B that used movement in the sagittal plane could not decrease RMSE values in many attachment conditions. Performing the movement in the sagittal plane was considered to be difficult for practical use.
{"title":"A Basic Test of Calibration Methods for Measurement of Three-Dimensional Movements of Lower Limbs with Inertial Sensors","authors":"Takuma Ando, Takashi Watanabe","doi":"10.1109/LSC.2018.8572202","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572202","url":null,"abstract":"Measurement of three-dimensional lower limb joint angles are useful to evaluate changes of movements after various lower limb diseases or injuries. However, estimation of three-dimensional angles with inertial measurement units (IMUs) causes errors, especially in abduction/adduction angle and in internal/external rotation angle. This was considered to be caused by difference between the sensor coordinate system and the body coordinate system. In order to solve the problem, various calibration methods of the coordinate system have been proposed. In this paper, three calibration methods, which were selected based on practical application, were examined in estimation of three-dimensional angles of a rigid body model under the 8 attachment conditions of IMU. The Method A determined the body coordinate system by measuring the vertical axis during a standing upright posture and another posture in the sagittal plane. The Method B estimated the sagittal plane and the normal vector of the plane from measured acceleration signals during movement of each segment of the lower limb in the sagittal plane. Method C was similar to Method A, in which postures of vertical and horizontal positions of lower limbs were used. Difference of the coordinate system of the IMU increased significantly RMSE values of estimated angles. Since the Method A and C showed almost same RMSE values as in the case that there was no difference of the coordinate system, the methods are considered to be practical. However, the Method B that used movement in the sagittal plane could not decrease RMSE values in many attachment conditions. Performing the movement in the sagittal plane was considered to be difficult for practical use.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"10 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132933970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572154
S. Rosati, G. Balestra, P. Franco, C. Fiandra, F. Arcadipane, P. Silvetti, U. Ricardi, E. Gallio
The radiation dose received by the pelvic Bone Marrow (BM) is a predictive factor for Hematologic Toxicity (HT) occurrence in the treatment of anal cancer. For this reason it is important to avoid BM during radiotherapy. In particular, the standard strategy in these cases consists in the identification of hematopoietically active BM (actBM), i.e. the part of BM in charge of blood cells generation, on 18FDG-PET, FLT-PET or MRI, but no approached have been developed for identifying actBM from CT images. This exploratory study aims to use radiomics for detecting actBM on CT sequences. Our approach is based on the extraction of 36 first-order and texture (second-order) features for each CT slice. These features are used as input of a Decision Tree (DT) classifier able to discriminate between active and inactive BM regions on the images. This method was applied to five patients affected by carcinoma of the anal canal and the obtained actBM segmentation was compared with the standard actBM identification from 18FDG-PET (reference standard, RS). Our results show that actBM identification in lumbosacral and iliac structures using radiomics overlaps the RS for more than 75% in 4 out of 5 patients.
{"title":"Radiomics for Identification of Active Bone Marrow from CT: An Exploratory Study","authors":"S. Rosati, G. Balestra, P. Franco, C. Fiandra, F. Arcadipane, P. Silvetti, U. Ricardi, E. Gallio","doi":"10.1109/LSC.2018.8572154","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572154","url":null,"abstract":"The radiation dose received by the pelvic Bone Marrow (BM) is a predictive factor for Hematologic Toxicity (HT) occurrence in the treatment of anal cancer. For this reason it is important to avoid BM during radiotherapy. In particular, the standard strategy in these cases consists in the identification of hematopoietically active BM (actBM), i.e. the part of BM in charge of blood cells generation, on 18FDG-PET, FLT-PET or MRI, but no approached have been developed for identifying actBM from CT images. This exploratory study aims to use radiomics for detecting actBM on CT sequences. Our approach is based on the extraction of 36 first-order and texture (second-order) features for each CT slice. These features are used as input of a Decision Tree (DT) classifier able to discriminate between active and inactive BM regions on the images. This method was applied to five patients affected by carcinoma of the anal canal and the obtained actBM segmentation was compared with the standard actBM identification from 18FDG-PET (reference standard, RS). Our results show that actBM identification in lumbosacral and iliac structures using radiomics overlaps the RS for more than 75% in 4 out of 5 patients.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131389702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/LSC.2018.8572220
Narges Hossein-Zadeh, M. Daliri, S. Magierowski, E. Ghafar-Zadeh
The realization of miniaturized Nuclear Magnetic Resonance (NMR) technology has received significant attention from researchers in both industry and academia. In this paper, we take a step toward the development of a fully integrated NMR by addressing the challenge of background magnetic resonance (MR) signal cancellation. A new fully differential oscillator-based NMR transceiver is proposed. This topology can suppress the background MR signal and enhance the sensitivity of the NMR transceiver. The proposed circuit contains a LC-Tank oscillator incorporated with a variable gain amplifier (VGA). This NMR transceiver is designed at 21 MHz NMR frequency. Post-layout simulations of the integrated circuit were performed using $0.18-mu mathbf{m}$ CMOS technology. These results prove the functionality and applicability of the proposed circuit for NMR applications using a commercially available 0.5-Tesla magnet.
{"title":"A Novel Fully Differential NMR Transciever","authors":"Narges Hossein-Zadeh, M. Daliri, S. Magierowski, E. Ghafar-Zadeh","doi":"10.1109/LSC.2018.8572220","DOIUrl":"https://doi.org/10.1109/LSC.2018.8572220","url":null,"abstract":"The realization of miniaturized Nuclear Magnetic Resonance (NMR) technology has received significant attention from researchers in both industry and academia. In this paper, we take a step toward the development of a fully integrated NMR by addressing the challenge of background magnetic resonance (MR) signal cancellation. A new fully differential oscillator-based NMR transceiver is proposed. This topology can suppress the background MR signal and enhance the sensitivity of the NMR transceiver. The proposed circuit contains a LC-Tank oscillator incorporated with a variable gain amplifier (VGA). This NMR transceiver is designed at 21 MHz NMR frequency. Post-layout simulations of the integrated circuit were performed using $0.18-mu mathbf{m}$ CMOS technology. These results prove the functionality and applicability of the proposed circuit for NMR applications using a commercially available 0.5-Tesla magnet.","PeriodicalId":254835,"journal":{"name":"2018 IEEE Life Sciences Conference (LSC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114259790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：