Computer-aided autism diagnosis using visual attention models and eye-tracking: replication and improvement proposal.

BACKGROUND: Autism Spectrum Disorder (ASD) diagnosis can be aided by approaches based on eye-tracking signals. Recently, the feasibility of building Visual Attention Models (VAMs) from features extracted from visual stimuli and their use for classifying cases and controls has been demonstrated using Neural Networks and Support Vector Machines. The present work has three aims: 1) to evaluate whether the trained classifier from the previous study was generalist enough to classify new samples with a new stimulus; 2) to replicate the previously approach to train a new classifier with a new dataset; 3) to evaluate the performance of classifiers obtained by a new classification algorithm (Random Forest) using the previous and the current datasets. METHODS: The previously approach was replicated with a new stimulus and new sample, 44 from the Typical Development group and 33 from the ASD group. After the replication, Random Forest classifier was tested to substitute Neural Networks algorithm. RESULTS: The test with the trained classifier reached an AUC of 0.56, suggesting that the trained classifier requires retraining of the VAMs when changing the stimulus. The replication results reached an AUC of 0.71, indicating the potential of generalization of the approach for aiding ASD diagnosis, as long as the stimulus is similar to the originally proposed. The results achieved with Random Forest were superior to those achieved with the original approach, with an average AUC of 0.95 for the previous dataset and 0.74 for the new dataset. CONCLUSION: In summary, the results of the replication experiment were satisfactory, which suggests the robustness of the approach and the VAM-based approaches feasibility to aid in ASD diagnosis. The proposed method change improved the classification performance. Some limitations are discussed and additional studies are encouraged to test other conditions and scenarios.

Pathological beige remodeling induced by cancer cachexia depends on the disease severity and involves mainly the trans-differentiation of mature white adipocytes.

In cancer associated cachexia (CAC), white adipose tissue undergoes morphofunctional and inflammatory changes that lead to tissue dysfunction and remodeling. In addition to metabolic changes in white adipose tissues (WAT), adipose tissue atrophy has been implicated in several clinical complications and poor prognoses associated with cachexia. Adipocyte atrophy may be associated with increased beige remodeling in human CAC as evidenced by the "beige remodeling" observed in preclinical models of CAC. Even though beige remodeling is associated with CAC-induced WAT dysfunction, there are still some open questions regarding their cellular origins. In this study, we investigated the development of beige remodeling in CAC from a broader perspective. In addition, we used a grading system to identify the scAT as being affected by mice weight loss early and intensely. Using different in vitro and ex-vivo techniques, we demonstrated that Lewis LLC1 cells can induce a switch from white to beige adipocytes, which is specific to this type of tumor cell. During the more advanced stages of CAC, beige adipocytes are mainly formed from the transdifferentiation of cells. According to our results, humanizing the CAC classification system is an efficient approach to defining the onset of the syndrome in a more homogeneous manner. Pathological beige remodeling occurred early in the disease course and exhibited phenotypic characteristics specific to LLC cells' secretomes. Developing therapeutic strategies that recruit beige adipocytes in vivo may be better guided by an understanding of the cellular origins of beige adipocytes emitted by CAC.

Computer-aided autism diagnosis based on visual attention models using eye tracking.

An advantage of using eye tracking for diagnosis is that it is non-invasive and can be performed in individuals with different functional levels and ages. Computer/aided diagnosis using eye tracking data is commonly based on eye fixation points in some regions of interest (ROI) in an image. However, besides the need for every ROI demarcation in each image or video frame used in the experiment, the diversity of visual features contained in each ROI may compromise the characterization of visual attention in each group (case or control) and consequent diagnosis accuracy. Although some approaches use eye tracking signals for aiding diagnosis, it is still a challenge to identify frames of interest when videos are used as stimuli and to select relevant characteristics extracted from the videos. This is mainly observed in applications for autism spectrum disorder (ASD) diagnosis. To address these issues, the present paper proposes: (1) a computational method, integrating concepts of Visual Attention Model, Image Processing and Artificial Intelligence techniques for learning a model for each group (case and control) using eye tracking data, and (2) a supervised classifier that, using the learned models, performs the diagnosis. Although this approach is not disorder-specific, it was tested in the context of ASD diagnosis, obtaining an average of precision, recall and specificity of 90%, 69% and 93%, respectively.

Toll-Like Receptor-4 Disruption Suppresses Adipose Tissue Remodeling and Increases Survival in Cancer Cachexia Syndrome.

Cancer-induced cachexia, characterized by systemic inflammation, body weight loss, adipose tissue (AT) remodeling and muscle wasting, is a malignant metabolic syndrome with undefined etiology. Here, we show that both genetic ablation and pharmacological inhibition of TLR4 were able to attenuate the main clinical markers of cachexia in mice bearing Lewis lung carcinoma (LLC). AT remodelling was not found in LLC tumor-bearing (TB) TLR4-/- mice due to reduced macrophage infiltration and adipocyte atrophy. TLR4-/- mice were also resistant to cold-induced browning of subcutaneous AT (scAT). Importantly, pharmacological inhibition of TLR4 (Atorvastatin) reproduced the main protective effect against AT remodeling found in TLR4-/- TB mice. Moreover, the treatment was effective in prolonging survival and attenuating tumor mass growth when compared to non-treated-TB animals. Furthermore, tumor-induced elevation of circulating pro-inflammatory cytokines was similarly abolished in both genetic ablation and pharmacological inhibition of TLR4. These data suggest that TLR4 is a critical mediator and a promising target for novel anti-cachexia therapies.

Pioglitazone treatment increases survival and prevents body weight loss in tumor-bearing animals: possible anti-cachectic effect.

Cachexia is a multifactorial syndrome characterized by profound involuntary weight loss, fat depletion, skeletal muscle wasting, and asthenia; all symptoms are not entirely attributable to inadequate nutritional intake. Adipose tissue and skeletal muscle loss during cancer cachexia development has been described systematically. The former was proposed to precede and be more rapid than the latter, which presents a means for the early detection of cachexia in cancer patients. Recently, pioglitazone (PGZ) was proposed to exhibit anti-cancer properties, including a reduction in insulin resistance and adipose tissue loss; nevertheless, few studies have evaluated its effect on survival. For greater insight into a potential anti-cachectic effect due to PGZ, 8-week-old male Wistar rats were subcutaneously inoculated with 1 mL (2×107) of Walker 256 tumor cells. The animals were randomly assigned to two experimental groups: TC (tumor + saline-control) and TP5 (tumor + PGZ/5 mg). Body weight, food ingestion and tumor growth were measured at baseline and after removal of tumor on days 7, 14 and 26. Samples from different visceral adipose tissue (AT) depots were collected on days 7 and 14 and stored at -80o C (5 to 7 animals per day/group). The PGZ treatment showed an increase in the survival average of 27.3% (P< 0.01) when compared to TC. It was also associated with enhanced body mass preservation (40.7 and 56.3%, p< 0.01) on day 14 and 26 compared with the TC group. The treatment also reduced the final tumor mass (53.4%, p<0.05) and anorexia compared with the TC group during late-stage cachexia. The retroperitoneal AT (RPAT) mass was preserved on day 7 compared with the TC group during the same experimental period. Such effect also demonstrates inverse relationship with tumor growth, on day 14. Gene expression of PPAR-γ, adiponectin, LPL and C/EBP-α from cachectic rats was upregulated after PGZ. Glucose uptake from adipocyte cells (RPAT) was entirely re-established due to PGZ treatment. Taken together, the results demonstrate beneficial effects of PGZ treatment at both the early and final stages of cachexia.