Rashid Ansari - Academia.edu (original) (raw)
Papers by Rashid Ansari
ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
We present a novel deep learning method for fully automated detection and classification of the C... more We present a novel deep learning method for fully automated detection and classification of the Cervical Vertebrae Maturation (CVM) stages. The deep convolutional neural network consists of three parallel networks (TriPodNet) independently trained with different initialization parameters. They also have a built-in set of novel directional filters that highlight the Cervical Vertebrae edges in X-ray images. Outputs of the three parallel networks are combined using a fully connected layer. 1018 cephalometric radiographs were labeled, divided by gender, and classified according to the CVM stages. Resulting images, using different training techniques and patches, were used to train TripodNet together with a set of tunable directional edge enhancers. Data augmentation is implemented to avoid overfitting. TripodNet achieves the state-of-the-art accuracy of 81.18% in female patients and 75.32% in male patients. The proposed Tripod-Net achieves a higher accuracy in our dataset than the Swin Transformers and the previous network models that we investigated for CVM stage estimation.
ArXiv, 2019
Electroencephalography (EEG) serves as an effective diagnostic tool for mental disorders and neur... more Electroencephalography (EEG) serves as an effective diagnostic tool for mental disorders and neurological abnormalities. Enhanced analysis and classification of EEG signals can help improve detection performance. This work presents a new approach that seeks to exploit the knowledge of EEG sensor spatial configuration to achieve higher detection accuracy. Two classification models, one which ignores the configuration (model 1) and one that exploits it with different interpolation methods (model 2), are studied. The analysis is based on the information content of these signals represented in two different ways: concatenation of the channels of the frequency bands and an image-like 2D representation of the EEG channel locations. Performance of these models is examined on two tasks, social anxiety disorder (SAD) detection, and emotion recognition using DEAP dataset. Validity of our hypothesis that model 2 will significantly outperform model 1 is borne out in the results, with accuracy $...
Research in Gerontological Nursing, 2014
The nursing outcomes of hospitalized patients whose plans of care include death anxiety, which is... more The nursing outcomes of hospitalized patients whose plans of care include death anxiety, which is a diagnosis among patients at the end-of-life, are obscure. The authors of the current article applied data mining techniques to nursing plan-of-care data for patients diagnosed with death anxiety, as defined by North American Nursing Diagnosis Association International, from four different hospitals to examine nursing care outcomes and associated factors. Results indicate that <50% of patients met the expected outcome of comfortable death. Gerontology unit patients were more likely to meet the expected outcome than patients from other unit types, although results were not statistically significant. Younger patients (i.e., age <65) had a lower chance of meeting the outcome compared with older patients (i.e., age ≥65) (χ 2 (1) = 9.266, p < 0.004). Longer stays improved the chances of meeting the outcome (χ 2 (2) = 6.47, p < 0.04). Results indicate that death anxiety outcomes are suboptimal and suggest the need to better educate clinicians about diagnosing and treating death anxiety among patients who face the end-of-life transition. The official North American Nursing Diagnosis Association International (NANDA-I) definition for death anxiety is a "vague uneasy feeling of discomfort or dread generated by perceptions of a real or imagined threat to one's existence" (2007, pp. 11-12); it is characterized by fear of suffering, powerlessness, and the impact of one's death on others. Death anxiety has also been described as apprehension arising from one's death awareness (Abdel-Khalek, 2005
Lecture Notes in Computer Science, 2014
In this paper, we present a digital check forgery attack on check processing systems used in onli... more In this paper, we present a digital check forgery attack on check processing systems used in online banking that results in check fraud. Such an attack is facilitated by multiple factors: the use of digital images to perform check transactions, advances in image processing technologies, the use of untrusted client-side devices and software, and the modalities of deposit. We note that digital check forgery attacks offer better chances of success in committing fraud when compared with conventional check forgery attacks. We discuss an instance of this attack and find several leading banks vulnerable to digital check forgery.
This paper presents a novel blind watermark detection/decoding scheme for spread spectrum (SS) ba... more This paper presents a novel blind watermark detection/decoding scheme for spread spectrum (SS) based watermarking exploiting the fact that the embedded watermark and the host signal are mutually independent. The proposed detector assumes that the host signal and the watermark obey non-Gaussian distributions. The proposed scheme employs the theory of blind source separation (BSS) using independent component analysis (ICA) to cancel the host-signal interference at the detector/decoder. The paper presents analytical results showing that the proposed detector performs significantly better than the existing blind detectors used for SS-based watermarking. The paper also shows that the detection performance approaches to that of an informed detector as interference due to the host signal in the estimated watermark using BSS approaches to zero.
Learning from Knowledge Engineering, 2001
Cooperation among mobile devices and utilizing multiple interfaces such as cellular and local are... more Cooperation among mobile devices and utilizing multiple interfaces such as cellular and local area links simultaneously are promising to meet the increasing throughput demand over cellular links. In particular, when mobile devices are in the close proximity of each other and are interested in the same content, device-to-device connections such as WiFi-Direct, in addition to cellular links, can be utilized to construct a cooperative system. However, it is crucial to understand the potential of network coding for cooperating mobile devices with multiple interfaces. In this paper, we consider this problem, and (i) develop network coding schemes for cooperative mobile devices with multiple interfaces, and (ii) characterize the performance of network coding by using the number of transmissions to recover all packets as a performance metric.
Patient pain can be detected highly reliably from facial expressions using a set of facial muscle... more Patient pain can be detected highly reliably from facial expressions using a set of facial muscle-based action units (AUs) defined by the Facial Action Coding System (FACS). A key characteristic of facial expression of pain is the simultaneous occurrence of pain-related AU combinations, whose automated detection would be highly beneficial for efficient and practical pain monitoring. Existing general Automated Facial Expression Recognition (AFER) systems prove inadequate when applied specifically for detecting pain as they either focus on detecting individual pain-related AUs but not on combinations or they seek to bypass AU detection by training a binary pain classifier directly on pain intensity data but are limited by lack of enough labeled data for satisfactory training. In this paper, we propose a new approach that mimics the strategy of human coders of decoupling pain detection into two consecutive tasks: one performed at the individual video-frame level and the other at video-...
Publication in the conference proceedings of EUSIPCO, Antalya, Turkey, 2005
Publication in the conference proceedings of EUSIPCO, Antalya, Turkey, 2005
Publication in the conference proceedings of EUSIPCO, Toulouse, France, 2002
Proceedings., International Conference on Image Processing
Page 1. EFFECTS OF IMAGE PREPROCESSING/RESIZING ON DIAGNOSTIC QUALITY OF COMPRESSED MEDICAL IMAGE... more Page 1. EFFECTS OF IMAGE PREPROCESSING/RESIZING ON DIAGNOSTIC QUALITY OF COMPRESSED MEDICAL IMAGES Barry J. Sullivan Rashid Ansari Ameritech, 2000 W. Ameritech Center Dr. Hoffman Estates, IL 60196-1025 ...
Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429)
IEEE Transactions on Circuits and Systems, 1985
[1991] Proceedings of the 34th Midwest Symposium on Circuits and Systems
... E 2 and a E R. Here, 2 and 72 denote the set of integers and real numbers, respectively ... F... more ... E 2 and a E R. Here, 2 and 72 denote the set of integers and real numbers, respectively ... For IIR filters with right-sided impulse responses and ratio-nal transfer functions, linear phase solutions do not ... with their respec-tive orders of N,, + 1 and Na + 1. Instead of approximating A(z ...
2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698), 2003
We present an algorithm for real-time adaptive forward error correction (FEC) of MPEG-2 video str... more We present an algorithm for real-time adaptive forward error correction (FEC) of MPEG-2 video stream, encapsulated using real-time transport protocol (RTP) and delivered over best-effort networks. Our algorithm provides an efficient method to determine the allocation of redundancy to the MPEG-2 video stream. The redundancy is allocated such that the resulting estimated degradation density function for video (DDF) is uniformly distributed. A weight, which indicates the relative importance of RTP packets, together with the communication channel characteristics and FEC scheme are used to model the density function of the video stream and allow us to determine the allocation of FEC packets. The weight is based on the content of RTP packets in the video stream. Parameters extracted from the RTP header are used to determine the weights, so that the proposed algorithm can be implemented in real-time. In our simulations, we have relied on motion compensation and group of picture (GOP) data to determine the relative weights. Simulation results provided establish the significant improvement in performance based on our proposed approach to adaptive FEC.
Proceedings of 1st International Conference on Image Processing
Page 1. RELATIVE EFFECTS OF RESOLUTION AND QUANTIZATION ON THE QUALITY OF COMPRESSED MEDICAL IMAG... more Page 1. RELATIVE EFFECTS OF RESOLUTION AND QUANTIZATION ON THE QUALITY OF COMPRESSED MEDICAL IMAGES Barry J. Sullivan Kashid Ansari Ameritech Bellcore 2000 W. Ameritech Center Dr. Hoffman Estates, IL 60196-1025 b. s ull i van @ieee . org ...
In this article a traffic light recognition with status detection system is introduced. The syste... more In this article a traffic light recognition with status detection system is introduced. The system is evaluated from a generic point of view but the applications range from Intelligent Transportation System (ITS) to visual impaired and color vision deficiencies aid to safely cross streets. The algorithm is based on color segmentation in HSV color space. After that candidates reduction is performed using a pipeline approach to speed up the algorithm. Resulting candidates are input to feature extraction and support vector machine is then applied. For the training of the Support Vector Machine a database with images collected in Chicago is used. Unlike other works the purpose is to evaluate different performance according to the feature extraction. In particular HOG, HAAR and LBP features are compared. The purpose is also to create a database to be used from other researchers. The result is accurate and reliable provided that good quality images are input to the system. Keywords—Traffi...
Device-to-device (D2D) connections among mobile devices is promising to meet the increasing throu... more Device-to-device (D2D) connections among mobile devices is promising to meet the increasing throughput demand over cellular links. In particular, when mobile devices are in close proximity of each other and are interested in the same content, D2D connections such as Wi-Fi Direct can be opportunistically used to construct a cooperative networking system. In this setup, it is crucial to understand, quantify, and exploit the potential of cooperative mobile devices when D2D connections are used in conjunction with cellular. This paper presents our work on effectively utilizing network coding for cooperative mobile devices that have D2D and cellular connections. In particular, we first develop “Content -Aware Network Coding” algorithm for cooperative mobile devices utilizing cellular and D2D connections. The core idea behind content-aware network coding is to deliver high quality multimedia content to mobile devices by taking into account the importance of each packet. Then, we develop N...
ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
We present a novel deep learning method for fully automated detection and classification of the C... more We present a novel deep learning method for fully automated detection and classification of the Cervical Vertebrae Maturation (CVM) stages. The deep convolutional neural network consists of three parallel networks (TriPodNet) independently trained with different initialization parameters. They also have a built-in set of novel directional filters that highlight the Cervical Vertebrae edges in X-ray images. Outputs of the three parallel networks are combined using a fully connected layer. 1018 cephalometric radiographs were labeled, divided by gender, and classified according to the CVM stages. Resulting images, using different training techniques and patches, were used to train TripodNet together with a set of tunable directional edge enhancers. Data augmentation is implemented to avoid overfitting. TripodNet achieves the state-of-the-art accuracy of 81.18% in female patients and 75.32% in male patients. The proposed Tripod-Net achieves a higher accuracy in our dataset than the Swin Transformers and the previous network models that we investigated for CVM stage estimation.
ArXiv, 2019
Electroencephalography (EEG) serves as an effective diagnostic tool for mental disorders and neur... more Electroencephalography (EEG) serves as an effective diagnostic tool for mental disorders and neurological abnormalities. Enhanced analysis and classification of EEG signals can help improve detection performance. This work presents a new approach that seeks to exploit the knowledge of EEG sensor spatial configuration to achieve higher detection accuracy. Two classification models, one which ignores the configuration (model 1) and one that exploits it with different interpolation methods (model 2), are studied. The analysis is based on the information content of these signals represented in two different ways: concatenation of the channels of the frequency bands and an image-like 2D representation of the EEG channel locations. Performance of these models is examined on two tasks, social anxiety disorder (SAD) detection, and emotion recognition using DEAP dataset. Validity of our hypothesis that model 2 will significantly outperform model 1 is borne out in the results, with accuracy $...
Research in Gerontological Nursing, 2014
The nursing outcomes of hospitalized patients whose plans of care include death anxiety, which is... more The nursing outcomes of hospitalized patients whose plans of care include death anxiety, which is a diagnosis among patients at the end-of-life, are obscure. The authors of the current article applied data mining techniques to nursing plan-of-care data for patients diagnosed with death anxiety, as defined by North American Nursing Diagnosis Association International, from four different hospitals to examine nursing care outcomes and associated factors. Results indicate that <50% of patients met the expected outcome of comfortable death. Gerontology unit patients were more likely to meet the expected outcome than patients from other unit types, although results were not statistically significant. Younger patients (i.e., age <65) had a lower chance of meeting the outcome compared with older patients (i.e., age ≥65) (χ 2 (1) = 9.266, p < 0.004). Longer stays improved the chances of meeting the outcome (χ 2 (2) = 6.47, p < 0.04). Results indicate that death anxiety outcomes are suboptimal and suggest the need to better educate clinicians about diagnosing and treating death anxiety among patients who face the end-of-life transition. The official North American Nursing Diagnosis Association International (NANDA-I) definition for death anxiety is a "vague uneasy feeling of discomfort or dread generated by perceptions of a real or imagined threat to one's existence" (2007, pp. 11-12); it is characterized by fear of suffering, powerlessness, and the impact of one's death on others. Death anxiety has also been described as apprehension arising from one's death awareness (Abdel-Khalek, 2005
Lecture Notes in Computer Science, 2014
In this paper, we present a digital check forgery attack on check processing systems used in onli... more In this paper, we present a digital check forgery attack on check processing systems used in online banking that results in check fraud. Such an attack is facilitated by multiple factors: the use of digital images to perform check transactions, advances in image processing technologies, the use of untrusted client-side devices and software, and the modalities of deposit. We note that digital check forgery attacks offer better chances of success in committing fraud when compared with conventional check forgery attacks. We discuss an instance of this attack and find several leading banks vulnerable to digital check forgery.
This paper presents a novel blind watermark detection/decoding scheme for spread spectrum (SS) ba... more This paper presents a novel blind watermark detection/decoding scheme for spread spectrum (SS) based watermarking exploiting the fact that the embedded watermark and the host signal are mutually independent. The proposed detector assumes that the host signal and the watermark obey non-Gaussian distributions. The proposed scheme employs the theory of blind source separation (BSS) using independent component analysis (ICA) to cancel the host-signal interference at the detector/decoder. The paper presents analytical results showing that the proposed detector performs significantly better than the existing blind detectors used for SS-based watermarking. The paper also shows that the detection performance approaches to that of an informed detector as interference due to the host signal in the estimated watermark using BSS approaches to zero.
Learning from Knowledge Engineering, 2001
Cooperation among mobile devices and utilizing multiple interfaces such as cellular and local are... more Cooperation among mobile devices and utilizing multiple interfaces such as cellular and local area links simultaneously are promising to meet the increasing throughput demand over cellular links. In particular, when mobile devices are in the close proximity of each other and are interested in the same content, device-to-device connections such as WiFi-Direct, in addition to cellular links, can be utilized to construct a cooperative system. However, it is crucial to understand the potential of network coding for cooperating mobile devices with multiple interfaces. In this paper, we consider this problem, and (i) develop network coding schemes for cooperative mobile devices with multiple interfaces, and (ii) characterize the performance of network coding by using the number of transmissions to recover all packets as a performance metric.
Patient pain can be detected highly reliably from facial expressions using a set of facial muscle... more Patient pain can be detected highly reliably from facial expressions using a set of facial muscle-based action units (AUs) defined by the Facial Action Coding System (FACS). A key characteristic of facial expression of pain is the simultaneous occurrence of pain-related AU combinations, whose automated detection would be highly beneficial for efficient and practical pain monitoring. Existing general Automated Facial Expression Recognition (AFER) systems prove inadequate when applied specifically for detecting pain as they either focus on detecting individual pain-related AUs but not on combinations or they seek to bypass AU detection by training a binary pain classifier directly on pain intensity data but are limited by lack of enough labeled data for satisfactory training. In this paper, we propose a new approach that mimics the strategy of human coders of decoupling pain detection into two consecutive tasks: one performed at the individual video-frame level and the other at video-...
Publication in the conference proceedings of EUSIPCO, Antalya, Turkey, 2005
Publication in the conference proceedings of EUSIPCO, Antalya, Turkey, 2005
Publication in the conference proceedings of EUSIPCO, Toulouse, France, 2002
Proceedings., International Conference on Image Processing
Page 1. EFFECTS OF IMAGE PREPROCESSING/RESIZING ON DIAGNOSTIC QUALITY OF COMPRESSED MEDICAL IMAGE... more Page 1. EFFECTS OF IMAGE PREPROCESSING/RESIZING ON DIAGNOSTIC QUALITY OF COMPRESSED MEDICAL IMAGES Barry J. Sullivan Rashid Ansari Ameritech, 2000 W. Ameritech Center Dr. Hoffman Estates, IL 60196-1025 ...
Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429)
IEEE Transactions on Circuits and Systems, 1985
[1991] Proceedings of the 34th Midwest Symposium on Circuits and Systems
... E 2 and a E R. Here, 2 and 72 denote the set of integers and real numbers, respectively ... F... more ... E 2 and a E R. Here, 2 and 72 denote the set of integers and real numbers, respectively ... For IIR filters with right-sided impulse responses and ratio-nal transfer functions, linear phase solutions do not ... with their respec-tive orders of N,, + 1 and Na + 1. Instead of approximating A(z ...
2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698), 2003
We present an algorithm for real-time adaptive forward error correction (FEC) of MPEG-2 video str... more We present an algorithm for real-time adaptive forward error correction (FEC) of MPEG-2 video stream, encapsulated using real-time transport protocol (RTP) and delivered over best-effort networks. Our algorithm provides an efficient method to determine the allocation of redundancy to the MPEG-2 video stream. The redundancy is allocated such that the resulting estimated degradation density function for video (DDF) is uniformly distributed. A weight, which indicates the relative importance of RTP packets, together with the communication channel characteristics and FEC scheme are used to model the density function of the video stream and allow us to determine the allocation of FEC packets. The weight is based on the content of RTP packets in the video stream. Parameters extracted from the RTP header are used to determine the weights, so that the proposed algorithm can be implemented in real-time. In our simulations, we have relied on motion compensation and group of picture (GOP) data to determine the relative weights. Simulation results provided establish the significant improvement in performance based on our proposed approach to adaptive FEC.
Proceedings of 1st International Conference on Image Processing
Page 1. RELATIVE EFFECTS OF RESOLUTION AND QUANTIZATION ON THE QUALITY OF COMPRESSED MEDICAL IMAG... more Page 1. RELATIVE EFFECTS OF RESOLUTION AND QUANTIZATION ON THE QUALITY OF COMPRESSED MEDICAL IMAGES Barry J. Sullivan Kashid Ansari Ameritech Bellcore 2000 W. Ameritech Center Dr. Hoffman Estates, IL 60196-1025 b. s ull i van @ieee . org ...
In this article a traffic light recognition with status detection system is introduced. The syste... more In this article a traffic light recognition with status detection system is introduced. The system is evaluated from a generic point of view but the applications range from Intelligent Transportation System (ITS) to visual impaired and color vision deficiencies aid to safely cross streets. The algorithm is based on color segmentation in HSV color space. After that candidates reduction is performed using a pipeline approach to speed up the algorithm. Resulting candidates are input to feature extraction and support vector machine is then applied. For the training of the Support Vector Machine a database with images collected in Chicago is used. Unlike other works the purpose is to evaluate different performance according to the feature extraction. In particular HOG, HAAR and LBP features are compared. The purpose is also to create a database to be used from other researchers. The result is accurate and reliable provided that good quality images are input to the system. Keywords—Traffi...
Device-to-device (D2D) connections among mobile devices is promising to meet the increasing throu... more Device-to-device (D2D) connections among mobile devices is promising to meet the increasing throughput demand over cellular links. In particular, when mobile devices are in close proximity of each other and are interested in the same content, D2D connections such as Wi-Fi Direct can be opportunistically used to construct a cooperative networking system. In this setup, it is crucial to understand, quantify, and exploit the potential of cooperative mobile devices when D2D connections are used in conjunction with cellular. This paper presents our work on effectively utilizing network coding for cooperative mobile devices that have D2D and cellular connections. In particular, we first develop “Content -Aware Network Coding” algorithm for cooperative mobile devices utilizing cellular and D2D connections. The core idea behind content-aware network coding is to deliver high quality multimedia content to mobile devices by taking into account the importance of each packet. Then, we develop N...