The analysis of veterinary radiographic imaging data is an essential step in the diagnosis of many thoracic lesions. Given the limited time that physicians can devote to a single patient, it would be valuable to implement an automated system to help clinicians make faster but still accurate diagnoses. Currently, most of such systems are based on supervised deep learning approaches. However, the problem with these solutions is that they need a large database of labeled data. Access to such data is often limited, as it requires a great investment of both time and money. Therefore, in this work we present a solution that allows higher classification scores to be obtained using knowledge transfer from inter-species and inter-pathology self-supervised learning methods. Before training the network for classification, pretraining of the model was performed using self-supervised learning approaches on publicly available unlabeled radiographic data of human and dog images, which allowed substantially increasing the number of images for this phase. The self-supervised learning approaches included the Beta Variational Autoencoder, the Soft-Introspective Variational Autoencoder, and a Simple Framework for Contrastive Learning of Visual Representations. After the initial pretraining, fine-tuning was performed for the collected veterinary dataset using 20% of the available data. Next, a latent space exploration was performed for each model after which the encoding part of the model was fine-tuned again, this time in a supervised manner for classification. Simple Framework for Contrastive Learning of Visual Representations proved to be the most beneficial pretraining method. Therefore, it was for this method that experiments with various fine-tuning methods were carried out. We achieved a mean ROC AUC score of 0.77 and 0.66, respectively, for the laterolateral and dorsoventral projection datasets. The results show significant improvement compared to using the model without any pretraining approach.

Over the last few years, a large number of studies have been conducted on the monitoring of human behavior remaining beyond conscious control. One area of application for such monitoring systems is lie detection. The most popular method currently used for this purpose is polygraph examination, which has proven its usefulness in the field and in laboratories, but it is not without its drawbacks. Technological advances in data acquisition and automated analysis have ensured that contactless tools are in high demand in security fields like airport screening or pre-employment procedures. As a result, there has been a shift in interest away from traditional polygraph examinations toward the analysis of facial expressions, voice, and speech patterns, as well as eye-tracking signals to detect deceptive behavior. In this paper, we focus on the last aspect, offer a comprehensive overview of two distinct lie detection methodologies based on eye tracking, and examine the commonly used oculomotor feature analysis. Furthermore, we explore current research directions and their results within the context of their potential applications in the field of forensics. We also highlight future research prospects, suggesting the utilization of eye tracking and scan path interpretation methodologies as a potential fully functional alternative for the conventional polygraph in the future. These considerations refer to legal and ethical issues related to the use of new technology to detect lies.

The widespread of audio-visual online communication highlights the importance of enhancing their ability to transmit emotional color and personal experience. Non-verbal biometric cues and signals were recently recognized to convey such missing emotional context when added to virtual interactions. Motivated by that, we present a haptic system allowing for a biometric signal transfer in a fully touchless and seamless way. We utilize camera-based heart-rate signal readings and ultrasonic mid-air haptic technology to affect the audience with a temporal tactile pattern representing the speaker’s heart rate. We assess the usability and engagement enhancement of such a system in two user-studies involving one-way communication, i.e., watching a short emotional video. Our analysis of biometric data and subjective responses hints toward changed values of arousal and valence as well as physiological responses when the haptic feedback was applied in a group of participants. Finally, we outline a further research agenda to confirm our observations with different emotions and communication scenarios.

The detection of concealed knowledge related to a crime in forensic settings is usually performed using a traditional polygraph that records physiological responses to specific stimuli. While these reactive measures are standard, there is growing interest in non-contact behavioral markers that can identify deceptive intent. This study aimed to evaluate the efficacy of anticipatory eye-movement patterns and machine learning classifiers in discriminating between guilty and control participants during the pre-stimulus phase of the examination, identifying cognitive markers that emerge before visual stimuli are even presented. To investigate the effectiveness of eye-tracking measurements combined with machine learning in Concealed Information Test, we examined whether anticipatory eye-movement patterns can reliably distinguish between guilty and control participants. At this phase of examination, participants are not yet presented with visual stimuli. We identified features that could reveal subtle differences in cognitive load, arousal, and gaze control associated with the intent to deceive in a series of statistical tests. In our study, several oculomotor features, particularly scanpath-related measures and pupil diameter, were extracted to capture visual exploration patterns and attentional allocation during the anticipation phase of Concealed Information Test. The support vector machine classifier achieved a mean participant-level accuracy of 0.729, which was significantly higher than chance. These results suggest that anticipatory ocular behavior in CIT may provide valuable insights into cognitive processes associated with anticipation of visual test stimuli even before they are presented.

Can credibility assessment be based on methods more reliable than those currently used? Traditional polygraph examinations, although widely employed, have well-documented limitations, motivating the search for alternative, noninvasive approaches. In this study, we developed an eye-tracking-based method to detect concealed memory traces by analyzing patterns of eye movements during a Concealed Information Test. Unlike previous research that examined eye movements in isolation, we directly compared the performance of the eye-tracking method with polygraph measurements within the same participants and under identical experimental conditions. Key features extracted from eye movements were used in a machine learning classifier to distinguish between deceptive and truthful responses. The results show that our eye-tracking-based test can effectively discriminate between deceptive and truthful individuals, achieving an accuracy of 94.9%. In comparison, polygraph examinations accurately classified 88.2% of participants under the same conditions, though five recordings had to be excluded because the polygraph expert could not reach a definitive conclusion. Finally, this study also considers the advantages and limitations of the two approaches, as well as ethical concerns related to measuring involuntary responses and using “black box” scoring algorithms.