Element-specific determination of X-ray transmission signatures using neural networks

Forensic Science International

Day

et al. 2017

Self Cite

Blowflies (Diptera: Calliphoridae) are forensically important as they are known to be one of the first to colonise human remains. The larval stage is typically used to assist a forensic entomologists with adult flies rarely used as they are difficult to age because they remain morphologically similar once they have gone through the initial transformation upon hatching. However, being able to age them is of interest and importance within the field. This study examined the cuticular hydrocarbons (CHC) of Diptera: Calliphoridae species Lucilia sericata, Calliphora vicina and Calliphora vomitoria. The CHCs were extracted from the cuticles of adult flies and analysed using Gas Chromatography-Mass Spectrometry (GC-MS). The chemical profiles were examined for the two Calliphora species at intervals of day 1, 5, 10, 20 and 30 and up to day 10 for L. sericata. The results show significant chemical changes occurring between the immature and mature adult flies over the extraction period examined in this study. With the aid of a Principal Component Analysis (PCA) and Artificial Neural Networks (ANN), samples were seen to cluster, allowing for the age to be established within the aforementioned time frames. The use of ANNs allowed for the automatic classification of novel samples with very good performance. This was a proof of concept study, which developed a method allowing to age post-emergence adults by using their chemical profiles.

Section: Artificial Neural Networkmentioning

confidence: 99%

Adult fly age estimations using cuticular hydrocarbons and Artificial Neural Networks in forensically important Calliphoridae species

Moore

Forensic Science International

Day

et al. 2017

Self Cite

“…ANNs take inspiration from the workings of neurons in the biological brain, consisting of abstract and simplified versions of the networks that are found in their biological equivalent. Whilst interesting from a biological point of view, ANNs are also interesting from a pattern recognition perspective and have successfully been used in a wide variety of engineering domains, for example (Day et al., ; Sinha and Knight, ; Xu and Humar, ). One of the main attractions of ANNs in automated pattern recognition is their ability to learn from examples and apply this knowledge to novel data samples.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Defect Detection in Reinforced Concrete Using Random Neural Architectures

Computer aided Civil Eng

Day

Austin

et al. 2013

Detecting defects within reinforced concrete is vital to the safety and durability of our built infrastructure upon which we heavily rely. In this work a noninvasive technique, ElectroMagnetic Anomaly Detection (EMAD), is used which provides information into the electromagnetic properties of the reinforcing steel and for which data analysis is currently performed visually: an undesirable process. This article investigates the first use of two neural network approaches to automate the analysis of this data: Echo State Networks (ESNs) and Extreme Learning Machines (ELMs) where fast and efficient training procedures allow networks to be trained and evaluated in less time than traditional neural network approaches. Data collected from real-world concrete structures have been analyzed using these two approaches as well as using a simple threshold measure and a standard recurrent neural network. The ELM approach offers a significant improvement in performance for a single tendon-reinforced structure, while two ESN architectures provided best performance for a meshreinforced concrete structure. C 2013 Computer-Aided Civil and Infrastructure Engineering.

“…32 CHCs are found in the lipid wax layer of nearly all insects. They have different functions in 33 different species of insects and previous studies have shown their use for age estimation [1,[13][14][15][16], 34 sex [17] and species identification [18][19][20]. 35 The two main factors believed to be influential for the composition of hydrocarbon pools are 36 development/genetic factors and physiological state/environmental conditions [21][22][23].…”

mentioning

confidence: 99%

“…In this study, two 150 important SOM training parameters, the neighbourhood size and learning rate, were 151 updated during training after a set number of elapsed epochs as described in Day et al [33].…”

mentioning

confidence: 99%

Age estimation of Calliphora (Diptera: Calliphoridae) larvae using cuticular hydrocarbon analysis and Artificial Neural Networks

Moore

Forensic Science International

Adam

et al. 2016

Self Cite

Cuticular hydrocarbons were extracted daily from the larvae of two closely related blowflies Calliphora vicina and Calliphora vomitoria (Diptera:Calliphoridae). The hydrocarbons were then analysed using Gas Chromatography-Mass Spectrometry (GC-MS), with the aim of observing changes within their chemical profiles in order to determine the larval age. The hydrocarbons were examined daily for each species from 1 day old larvae until pupariation. The results show significant chemical changes occurring from the younger larvae to the post-feeding larvae. With the aid of a multivariate statistical method (Principal Component Analysis and Artificial Neural Networks), samples were clustered and classified, allowing for the larval age to be established. Results from this study allowed larvae to be aged to the day with at worst, 87% accuracy, which suggests there is great potential for the use of cuticular hydrocarbons present on larvae to give an indication of their age and hence potentially a valuable tool for minimum PMI estimations.