Evaluating artificial intelligence for comparative radiography.

INTRODUCTION: Comparative radiography is a forensic identification and shortlisting technique based on the comparison of skeletal structures in ante-mortem and post-mortem images. The images (e.g., 2D radiographs or 3D computed tomographies) are manually superimposed and visually compared by a forensic practitioner. It requires a significant amount of time per comparison, limiting its utility in large comparison scenarios. METHODS: We propose and validate a novel framework for automating the shortlisting of candidates using artificial intelligence. It is composed of (1) a segmentation method to delimit skeletal structures' silhouettes in radiographs, (2) a superposition method to generate the best simulated "radiographs" from 3D images according to the segmented radiographs, and (3) a decision-making method for shortlisting all candidates ranked according to a similarity metric. MATERIAL: The dataset is composed of 180 computed tomographies and 180 radiographs where the frontal sinuses are visible. Frontal sinuses are the skeletal structure analyzed due to their high individualization capability. RESULTS: Firstly, we validate two deep learning-based techniques for segmenting the frontal sinuses in radiographs, obtaining high-quality results. Secondly, we study the framework's shortlisting capability using both automatic segmentations and superimpositions. The obtained superimpositions, based only on the superimposition metric, allowed us to filter out 40% of the possible candidates in a completely automatic manner. Thirdly, we perform a reliability study by comparing 180 radiographs against 180 computed tomographies using manual segmentations. The results allowed us to filter out 73% of the possible candidates. Furthermore, the results are robust to inter- and intra-expert-related errors.

Left ventricular shape index and eccentricity index with ECG-gated Nitrogen-13 ammonia PET/CT in patients with myocardial infarction, ischemia, and normal perfusion.

BACKGROUND: LV geometry with shape index (SI) and eccentricity index (EI) measured by myocardial perfusion positron emission tomography/computed tomography (PET/CT) may allow the evaluation of left ventricular (LV) adverse remodeling. This first study aims to explore the relationship of SI and EI values acquired by Nitrogen-13 ammonia PET/CT in patients with normal perfusion, ischemia, and myocardial infarction. And evaluate the correlations between the variables of LV geometry, and with the variables of LV function. METHODS AND RESULTS: One hundred and forty patients who underwent an electrocardiogram (ECG)-gated PET/CT were selected and classified into 4 groups according to ischemia or infarction burden (normal perfusion, mild ischemia, moderate-severe ischemia, and infarction). The variables were automatically retrieved using dedicated software (QPS/QGS; Cedars-Sinai, Los Angeles, CA, USA). On multicomparison analysis (one-way ANOVA and Dunnett's Test), subjects in the infarction group had significant higher values of SI end-diastolic rest (P < 0.001), and stress (P = 0.003), SI end-systolic rest (P = 0.002) and stress (P < 0.001) as well as statistically significant lower values of EI rest (P < 0.001) and stress (P < 0.001) when compared with all other groups. Regarding Pearson correlation, in the infarcted group all the variables of SI and EI were significantly correlated (P < 0.001) with strong correlation coefficients (>0.60). SI end-systolic correlated significantly with the variables of LV function independently of the group of patients (P < 0.05). CONCLUSIONS: Shape and eccentricity indices differ in patients with myocardial infarction as compared to patients with ischemia or normal perfusion. This encourage further research in their potential for detecting LV adverse remodeling.

Computer-aided craniofacial superimposition validation study: the identification of the leaders and participants of the Polish-Lithuanian January Uprising (1863-1864).

In 2017, a series of human remains corresponding to the executed leaders of the "January Uprising" of 1863-1864 were uncovered at the Upper Castle of Vilnius (Lithuania). During the archeological excavations, 14 inhumation pits with the human remains of 21 individuals were found at the site. The subsequent identification process was carried out, including the analysis and cross-comparison of post-mortem data obtained in situ and in the lab with ante-mortem data obtained from historical archives. In parallel, three anthropologists with diverse backgrounds in craniofacial identification and two students without previous experience attempted to identify 11 of these 21 individuals using the craniofacial superimposition technique. To do this, the five participants had access to 18 3D scanned skulls and 14 photographs of 11 different candidates. The participants faced a cross-comparison problem involving 252 skull-face overlay scenarios. The methodology follows the main agreements of the European project MEPROCS and uses the software Skeleton-ID™. Based on MEPROCS standard, a final decision was provided within a scale, assigning a value in terms of strong, moderate, or limited support to the claim that the skull and the facial image belonged (or not) to the same person for each case. The problem of binary classification, positive/negative, with an identification rate for each participant was revealed. The results obtained in this study make the authors think that both the quality of the materials used and the previous experience of the analyst play a fundamental role when reaching conclusions using the CFS technique.

The clinical value of quantitative cardiovascular molecular imaging: a step towards precision medicine.

Cardiovascular diseases (CVD) are the leading cause of death worldwide and have an increasing impact on society. Precision medicine, in which optimal care is identified for an individual or a group of individuals rather than for the average population, might provide significant health benefits for this patient group and decrease CVD morbidity and mortality. Molecular imaging provides the opportunity to assess biological processes in individuals in addition to anatomical context provided by other imaging modalities and could prove to be essential in the implementation of precision medicine in CVD. New developments in single-photon emission computed tomography (SPECT) and positron emission tomography (PET) systems, combined with rapid innovations in promising and specific radiopharmaceuticals, provide an impressive improvement of diagnostic accuracy and therapy evaluation. This may result in improved health outcomes in CVD patients, thereby reducing societal impact. Furthermore, recent technical advances have led to new possibilities for accurate image quantification, dynamic imaging, and quantification of radiotracer kinetics. This potentially allows for better evaluation of disease activity over time and treatment response monitoring. However, the clinical implementation of these new methods has been slow. This review describes the recent advances in molecular imaging and the clinical value of quantitative PET and SPECT in various fields in cardiovascular molecular imaging, such as atherosclerosis, myocardial perfusion and ischemia, infiltrative cardiomyopathies, systemic vascular diseases, and infectious cardiovascular diseases. Moreover, the challenges that need to be overcome to achieve clinical translation are addressed, and future directions are provided.

Automatic landmark annotation in 3D surface scans of skulls: Methodological proposal and reliability study.

BACKGROUND AND OBJECTIVES: Craniometric landmarks are essential in many biomedical applications, such as morphometric analysis or forensic identification. The process of locating landmarks is usually a manual and slow task, highly influenced by fatigue, skills and the experience of the practitioner. Localization errors are propagated and magnified in subsequent steps, which can result in incorrect measurements or assumptions. Thereby, standardization, reliability and reproducibility lay the foundations for the necessary accuracy in subsequent measurements or anatomical analysis. In this paper, we present an automatic method to annotate 3D surface skull models taking into account anatomical and geometrical features. METHODS: The proposed method follows a hybrid structure where a deformable template is used to initialize the landmark positions. Then, a refinement stage is applied using prior anatomical knowledge to ensure a correct placement. Our proposal is validated over thirty 3D skull scans of male Caucasians, acquired by hand-held surface scanning, and a set of 58 craniometric landmarks. A statistical analysis was carried out to analyze the inter- and intra-observer variability of manual annotations and the automatic results, along with a visual assessment of the final results. RESULTS: Inter-observer errors show significant differences, which are reflected in the expert consensus used as reference. The average localization error was 2.19±1.5 mm when comparing the automatic landmarks to the reference location. The subsequent visual analysis confirmed the reliability of the refinement method for most landmarks. CONCLUSIONS: Repeated manual annotations show a high variability depending on both skills and expertise of the observer, and landmarks' location and characteristics. In contrast, the automatic method provides an accurate, robust and reproducible alternative to the tedious and error-prone task of manual landmarking.

Estimation of 2D to 3D dimensions and proportionality indices for facial examination.

Photo-anthropometry is a metric-based facial image comparison technique where measurements of the face are taken from an image using predetermined facial landmarks. In particular, dimensions and proportionality indices (DPIs) are compared to DPIs from another facial image. Different studies concluded that photo-anthropometric facial comparison, as it is currently practiced, is unsuitable for elimination purposes. The major limitation is the need for images acquired under very restrictive, controlled conditions. To overcome this latter issue, we propose a novel methodology to estimate 3D DPIs from 2D ones. It uses computer graphic techniques to simulate thousands of facial photographs under known camera conditions and regression to derive the mathematical relationship between 2D and 3D DPIs automatically. Additionally, we present a methodology that makes use of the estimated 3D DPIs for reducing the number of potential matches of a given unknown facial photograph within a set of known candidates. The error in the estimation of the 3D DPIs can be as large as 35%, but both I and III quartiles are consistently inside the ±5% range. The methodology for filtering cases has demonstrated to be useful in the task of narrowing down the list of possible candidates for a given photograph. It is able to remove on average (validated using cross-validation technique) 57% and 24% of the negative cases, depending on the amounts of DPIs available. Limitations of the work developed together with open research lines are included within the Discussion section.

Study on the criteria for assessing skull-face correspondence in craniofacial superimposition.

Craniofacial superimposition has the potential to be used as an identification method when other traditional biological techniques are not applicable due to insufficient quality or absence of ante-mortem and post-mortem data. Despite having been used in many countries as a method of inclusion and exclusion for over a century it lacks standards. Thus, the purpose of this research is to provide forensic practitioners with standard criteria for analysing skull-face relationships. Thirty-seven experts from 16 different institutions participated in this study, which consisted of evaluating 65 criteria for assessing skull-face anatomical consistency on a sample of 24 different skull-face superimpositions. An unbiased statistical analysis established the most objective and discriminative criteria. Results did not show strong associations, however, important insights to address lack of standards were provided. In addition, a novel methodology for understanding and standardizing identification methods based on the observation of morphological patterns has been proposed.

Past, present, and future of craniofacial superimposition: Literature and international surveys.

In this manuscript, the past, present and future of the identification of human remains based on craniofacial superimposition is reviewed. An analysis of the different technological approaches developed over time is offered in conjunction with a new classification based on the technology implemented throughout the diverse phases of the process. The state of the art of the technique, in the academic and forensic realms, is reflected in an extensive international survey that includes over one hundred experts worldwide. The results of the survey indicate the current relative importance of the technique, despite of its controversial nature within the scientific community. Finally, the future challenges to be faced to justify the use of this technique for either profiling, exclusion or identification purposes are discussed.

Artificial neuron-glia networks learning approach based on cooperative coevolution.

Artificial Neuron-Glia Networks (ANGNs) are a novel bio-inspired machine learning approach. They extend classical Artificial Neural Networks (ANNs) by incorporating recent findings and suppositions about the way information is processed by neural and astrocytic networks in the most evolved living organisms. Although ANGNs are not a consolidated method, their performance against the traditional approach, i.e. without artificial astrocytes, was already demonstrated on classification problems. However, the corresponding learning algorithms developed so far strongly depends on a set of glial parameters which are manually tuned for each specific problem. As a consequence, previous experimental tests have to be done in order to determine an adequate set of values, making such manual parameter configuration time-consuming, error-prone, biased and problem dependent. Thus, in this paper, we propose a novel learning approach for ANGNs that fully automates the learning process, and gives the possibility of testing any kind of reasonable parameter configuration for each specific problem. This new learning algorithm, based on coevolutionary genetic algorithms, is able to properly learn all the ANGNs parameters. Its performance is tested on five classification problems achieving significantly better results than ANGN and competitive results with ANN approaches.

Artificial astrocytes improve neural network performance.

Compelling evidence indicates the existence of bidirectional communication between astrocytes and neurons. Astrocytes, a type of glial cells classically considered to be passive supportive cells, have been recently demonstrated to be actively involved in the processing and regulation of synaptic information, suggesting that brain function arises from the activity of neuron-glia networks. However, the actual impact of astrocytes in neural network function is largely unknown and its application in artificial intelligence remains untested. We have investigated the consequences of including artificial astrocytes, which present the biologically defined properties involved in astrocyte-neuron communication, on artificial neural network performance. Using connectionist systems and evolutionary algorithms, we have compared the performance of artificial neural networks (NN) and artificial neuron-glia networks (NGN) to solve classification problems. We show that the degree of success of NGN is superior to NN. Analysis of performances of NN with different number of neurons or different architectures indicate that the effects of NGN cannot be accounted for an increased number of network elements, but rather they are specifically due to astrocytes. Furthermore, the relative efficacy of NGN vs. NN increases as the complexity of the network increases. These results indicate that artificial astrocytes improve neural network performance, and established the concept of Artificial Neuron-Glia Networks, which represents a novel concept in Artificial Intelligence with implications in computational science as well as in the understanding of brain function.

Brotes epizoóticos de triquinosis en dos criaderos de cerdos de la Región Metropolitana, Chile / Epizootic outbreaks of trichinosis in 2 pig farms from the Metropolitan Region, Chile

In Chile swine trichinosis has presented a progressive decreasing in the last two decades of XX century. T. spiralis pig infection descended from an average of 0,683 per 1000 in 1980-1984 to 0,315 in 1985-1989 and to 0,115 in 1990-1996. In the particular case of Metropolitan Region this decreasing has been more marked: from an average of 0,058 per 1000 in 1990-1994 to 0,003 in 1995-1999. Between the end of june 1999 and middle january 2000 in Metropolitan Region abattoirs T. spiralis was detected in 15 (4,9 percent) out of 306 swine from two pigsties located in El Monte (E.M) and Padre Hurtado (P.H) 45 and 30 km south-west from Santiago. In the same period another four pigs from the same premises were found infected in abattoirs of other regions. During inspection visits it was stated that both pig farms had deficient sanitary conditions. Phototrichinoscopy was positive in three out of five rattus norvegicus collected in E.M. In pigsty PH the examination of diaphragm samples of 25 dogs and 17 cats resulted negative. In the premises originating T. spirali infected swine the Metropolitan Environmetal Health Service Abattoirs Program carries out and epidemiological vigilance consisting in the follow-up of animls destined for slaughtering in order to initiate prophylactic actions oriented to eliminate eventual sources of trichinosis infection for human and rearing pigs