Derivation of an observer model adapted to irregular signals based on convolution channels.

Anthropomorphic model observers are mathe- matical algorithms which are applied to images with the ultimate goal of predicting human signal detection and classification accuracy across varieties of backgrounds, image acquisitions and display conditions. A limitation of current channelized model observers is their inability to handle irregularly-shaped signals, which are common in clinical images, without a high number of directional channels. Here, we derive a new linear model observer based on convolution channels which we refer to as the "Filtered Channel observer" (FCO), as an extension of the channelized Hotelling observer (CHO) and the nonprewhitening with an eye filter (NPWE) observer. In analogy to the CHO, this linear model observer can take the form of a single template with an external noise term. To compare with human observers, we tested signals with irregular and asymmetrical shapes spanning the size of lesions down to those of microcalfications in 4-AFC breast tomosynthesis detection tasks, with three different contrasts for each case. Whereas humans uniformly outperformed conventional CHOs, the FCO observer outperformed humans for every signal with only one exception. Additive internal noise in the models allowed us to degrade model performance and match human performance. We could not match all the human performances with a model with a single internal noise component for all signal shape, size and contrast conditions. This suggests that either the internal noise might vary across signals or that the model cannot entirely capture the human detection strategy. However, the FCO model offers an efficient way to apprehend human observer performance for a non-symmetric signal.

Study of criteria influencing the success rate of DNA swabs in operational conditions: a contribution to an evidence-based approach to crime scene investigation and triage

DNA is nowadays swabbed routinely to investigate serious and volume crimes, but research remains scarce when it comes to determining the criteria that may impact the success rate of DNA swabs taken on different surfaces and situations. To investigate these criteria in fully operational conditions, DNA analysis results of 4772 swabs taken by the forensic unit of a police department in Western Switzerland over a 2.5-year period (2012-2014) in volume crime cases were considered. A representative and random sample of 1236 swab analyses was extensively examined and codified, describing several criteria such as whether the swabbing was performed at the scene or in the lab, the zone of the scene where it was performed, the kind of object or surface that was swabbed, whether the target specimen was a touch surface or a biological fluid, and whether the swab targeted a single surface or combined different surfaces. The impact of each criterion and of their combination was assessed in regard to the success rate of DNA analysis, measured through the quality of the resulting profile, and whether the profile resulted in a hit in the national database or not. Results show that some situations - such as swabs taken on door and window handles for instance - have a higher success rate than average swabs. Conversely, other situations lead to a marked decrease in the success rate, which should discourage further analyses of such swabs. Results also confirm that targeting a DNA swab on a single surface is preferable to swabbing different surfaces with the intent to aggregate cells deposited by the offender. Such results assist in predicting the chance that the analysis of a swab taken in a given situation will lead to a positive result. The study could therefore inform an evidence-based approach to decision-making at the crime scene (what to swab or not) and at the triage step (what to analyse or not), contributing thus to save resource and increase the efficiency of forensic science efforts.

Exploring the relative DNA contribution of first and second object's users on mock touch DNA mixtures

Contact stains recovered at break-in crime scenes are frequently characterized by mixtures of DNA from several persons. Broad knowledge on the relative contribution of DNA left behind by different users overtime is of paramount importance. Such information might help crime investigators to robustly evaluate the possibility of detecting a specific (or known) individual's DNA profile based on the type and history of an object. To address this issue, a contact stain simulation-based protocol was designed. Fourteen volunteers either acting as first or second object's users were recruited. The first user was required to regularly handle/wear 9 different items during an 8-10-day period, whilst the second user for 5, 30 and 120 min, in three independent simulation sessions producing a total of 231 stains. Subsequently, the relative DNA profile contribution of each individual pair was investigated. Preliminary results showed a progressive increase of the percentage contribution of the second user compared to the first. Interestingly, the second user generally became the major DNA contributor when most objects were handled/worn for 120 min, Furthermore, the observation of unexpected additional alleles will then prompt the investigation of indirect DNA transfer events.

Probabilistic age classification with Bayesian networks: a study on the ossification status of the medial clavicular epiphysis

In the past few decades, the rise of criminal, civil and asylum cases involving young people lacking valid identification documents has generated an increase in the demand of age estimation. The chronological age or the probability that an individual is older or younger than a given age threshold are generally estimated by means of some statistical methods based on observations performed on specific physical attributes. Among these statistical methods, those developed in the Bayesian framework allow users to provide coherent and transparent assignments which fulfill forensic and medico-legal purposes. The application of the Bayesian approach is facilitated by using probabilistic graphical tools, such as Bayesian networks. The aim of this work is to test the performances of the Bayesian network for age estimation recently presented in scientific literature in classifying individuals as older or younger than 18 years of age. For these exploratory analyses, a sample related to the ossification status of the medial clavicular epiphysis available in scientific literature was used. Results obtained in the classification are promising: in the criminal context, the Bayesian network achieved, on the average, a rate of correct classifications of approximatively 97%, whilst in the civil context, the rate is, on the average, close to the 88%. These results encourage the continuation of the development and the testing of the method in order to support its practical application in casework.

Development of a systematic computer vision-based method to analyse and compare images of false identity documents for forensic intelligence purposes-Part I: acquisition, calibration and validation issues

Following their detection and seizure by police and border guard authorities, false identity and travel documents are usually scanned, producing digital images. This research investigates the potential of these images to classify false identity documents, highlight links between documents produced by a same modus operandi or same source, and thus support forensic intelligence efforts. Inspired by previous research work about digital images of Ecstasy tablets, a systematic and complete method has been developed to acquire, collect, process and compare images of false identity documents. This first part of the article highlights the critical steps of the method and the development of a prototype that processes regions of interest extracted from images. Acquisition conditions have been fine-tuned in order to optimise reproducibility and comparability of images. Different filters and comparison metrics have been evaluated and the performance of the method has been assessed using two calibration and validation sets of documents, made up of 101 Italian driving licenses and 96 Portuguese passports seized in Switzerland, among which some were known to come from common sources. Results indicate that the use of Hue and Edge filters or their combination to extract profiles from images, and then the comparison of profiles with a Canberra distance-based metric provides the most accurate classification of documents. The method appears also to be quick, efficient and inexpensive. It can be easily operated from remote locations and shared amongst different organisations, which makes it very convenient for future operational applications. The method could serve as a first fast triage method that may help target more resource-intensive profiling methods (based on a visual, physical or chemical examination of documents for instance). Its contribution to forensic intelligence and its application to several sets of false identity documents seized by police and border guards will be developed in a forthcoming article (part II).

Most and Least Preferred Colours Differ According to Object Context : New Insights from an Unrestricted Colour Range

Humans like some colours and dislike others, but which particular colours and why remains to be understood. Empirical studies on colour preferences generally targeted most preferred colours, but rarely least preferred (disliked) colours. In addition, findings are often based on general colour preferences leaving open the question whether results generalise to specific objects. Here, 88 participants selected the colours they preferred most and least for three context conditions (general, interior walls, t-shirt) using a high-precision colour picker. Participants also indicated whether they associated their colour choice to a valenced object or concept. The chosen colours varied widely between individuals and contexts and so did the reasons for their choices. Consistent patterns also emerged, as most preferred colours in general were more chromatic, while for walls they were lighter and for t-shirts they were darker and less chromatic compared to least preferred colours. This meant that general colour preferences could not explain object specific colour preferences. Measures of the selection process further revealed that, compared to most preferred colours, least preferred colours were chosen more quickly and were less often linked to valenced objects or concepts. The high intra- and inter-individual variability in this and previous reports furthers our understanding that colour preferences are determined by subjective experiences and that most and least preferred colours are not processed equally.