| 1. |
|
|
| 2. |
- Awad, Ali Ismail, et al.
(författare)
-
Impact of Some Biometric Modalities on Forensic Science
- 2014
-
Ingår i: Computational Intelligence in Digital Forensics: Forensic Investigation and Applications. - Cham : Springer International Publishing. - 9783319058849 - 9783319058856 ; , s. 47-62
-
Bokkapitel (refereegranskat)abstract
- Recently, forensic science has had many challenges in many different types of crimes and crime scenes, vary from physical crimes to cyber or computer crimes. Accurate and efficient human identification or recognition have become crucial for forensic applications due to the large diversity of crime scenes, and because of the increasing need to accurately identify criminals from the available crime evidences. Biometrics is an emerging technology that provides accurate and highly secure personal identification and verification systems for civilian and forensic applications. The positive impact of biometric modalities on forensic science began with the rapid developments in computer science, computational intelligence, and computing approaches. These advancements have been reflected in the biometric modality capturing process, feature extraction, feature robustness, and features matching. A complete and automatic biometric identification or recognition systems have been built accordingly. This chapter presents a study of the impacts of using some biometric modalities in forensic applications. Although biometrics identification replaces human work with computerized and automatic systems in order to achieve better performance, new challenges have arisen. These challenges lie in biometric system reliability and accuracy, system response time, data mining and classification, and protecting user privacy. This chapter sheds light on the positive and the negative impacts of using some biometric modalities in forensic science. In particular, the impacts of fingerprint image, facial image, and iris patterns are considered. The selected modalities are covered preliminarily before tackling their impact on forensic applications. Furthermore, an extensive look at the future of biometric modalities deployment in forensic applications is covered as the last part of the chapter.
|
|
| 3. |
|
|
| 4. |
- Elghamrawy, Sally M., et al.
(författare)
-
Genetic-based adaptive momentum estimation for predicting mortality risk factors for COVID-19 patients using deep learning
- 2022
-
Ingår i: International journal of imaging systems and technology (Print). - : John Wiley & Sons. - 0899-9457 .- 1098-1098. ; 32:2, s. 614-628
-
Tidskriftsartikel (refereegranskat)abstract
- The mortality risk factors for coronavirus disease (COVID-19) must be early predicted, especially for severe cases, to provide intensive care before they develop to critically ill immediately. This paper aims to develop an optimized convolution neural network (CNN) for predicting mortality risk factors for COVID-19 patients. The proposed model supports two types of input data clinical variables and the computed tomography (CT) scans. The features are extracted from the optimized CNN phase and then applied to the classification phase. The CNN model's hyperparameters were optimized using a proposed genetic-based adaptive momentum estimation (GB-ADAM) algorithm. The GB-ADAM algorithm employs the genetic algorithm (GA) to optimize Adam optimizer's configuration parameters, consequently improving the classification accuracy. The model is validated using three recent cohorts from New York, Mexico, and Wuhan, consisting of 3055, 7497,504 patients, respectively. The results indicated that the most significant mortality risk factors are: CD 8+ T Lymphocyte (Count), D-dimer greater than 1 Ug/ml, high values of lactate dehydrogenase (LDH), C-reactive protein (CRP), hypertension, and diabetes. Early identification of these factors would help the clinicians in providing immediate care. The results also show that the most frequent COVID-19 signs in CT scans included ground-glass opacity (GGO), followed by crazy-paving pattern, consolidations, and the number of lobes. Moreover, the experimental results show encouraging performance for the proposed model compared with different predicting models.
|
|
