Background: Early diagnosis of skin cancer lesions by dermoscopy, the gold standard in dermatological imaging, calls for a diagnostic upscale. The aim of the study was to improve the accuracy of dermoscopic skin cancer diagnosis through use of novel deep learning (DL) algorithms. An additional sonification-derived diagnostic layer was added to the visual classification to increase sensitivity. Methods: Two parallel studies were conducted: a laboratory retrospective study (LABS, n = 482 biopsies) and a non-interventional prospective observational study (OBS, n = 63 biopsies). A training data set of biopsy-verified reports, normal and cancerous skin lesions (n = 3954), were used to develop a DL classifier exploring visual features (System A). The outputs of the classifier were sonified, i.e. data conversion into sound (System B). Derived sound files were analyzed by a second machine learning classifier, either as raw audio (LABS, OBS) or following conversion into spectrograms (LABS) and by image analysis and human heuristics (OBS). The OBS criteria outcomes were System A specificity and System B sensitivity as raw sounds, spectrogram areas or heuristics. Findings: LABS employed dermoscopies, half benign half malignant, and compared the accuracy of Systems A and B. System A algorithm resulted in a ROC AUC of 0.976 (95% CI, 0.965-0.987). Secondary machine learning analysis of raw sound, FFT and Spectrogram ROC curves resulted in AUC's of 0.931 (95% CI 0.881-0.981), 0.90 (95% CI 0.838-0.963) and 0.988 (CI 95% 0.973-1.001), respectively. OBS analysis of raw sound dermoscopies by the secondary machine learning resulted in a ROC AUC of 0.819 (95% CI, 0.7956 to 0.8406). OBS image analysis of AUC for spectrograms displayed a ROC AUC of 0.808 (CI 95% 0.6945 To 0.9208). By applying a heuristic analysis of Systems A and B a sensitivity of 86% and specificity of 91% were derived in the clinical study. Interpretation: Adding a second stage of processing, which includes a deep learning algorithm of sonification and heuristic inspection with machine learning, significantly improves diagnostic accuracy. A combined two-stage system is expected to assist clinical decisions and de-escalate the current trend of over-diagnosis of skin cancer lesions as pathological.

The oblique parameters S, T, and U and their higher-order extensions (V, W, and X) are observables that combine electroweak precision data to quantify deviation from the Standard Model. These parameters were calculated at one loop in the basis-independent CP-violating Two-Higgs Doublet Model (2HDM). The scalar parameter space of the 2HDM was randomly sampled within limits imposed by unitarity and found to produce values of the oblique parameters within experimental bounds, with the exception of T. The experimental limits on T were used to predict information about the mass of the charged Higgs boson and the difference in mass between the charged Higgs boson and the heaviest neutral Higgs boson. In particular, it was found that the 2HDM predicts -600 GeV < m H ± − m3 <100 GeV, with values of m H ± > 250 GeV being preferred. The mass scale of the new physics (MNP ) produced by random sampling was consistently fairly high, with the average of the scalar masses falling between 400 and 800 GeV for Y2 = m 2 W , although the model can be tuned to produce a light neutral Higgs mass (∼ 120 GeV). Hence, the values produced for V, W, and X fell well within .01 of zero, confirming the robustness of the linear expansion approximation. Taking the CP-conserving limit of the model was found to not significantly affect the values generated for the oblique parameters. *

Abstract:We used molecular dynamics simulations, the Green-Kubo Modal Analysis (GKMA) method and sonification to study the modal contributions to thermal conductivity in individual polythiophene chains. The simulations suggest that it is possible to achieve divergent thermal conductivity and the GKMA method allowed for exact pinpointing of the modes responsible for the anomalous behavior. The analysis showed that transverse vibrations in the plane of the aromatic rings at low frequencies ~ 0.05 THz are primarily responsible. Further investigation showed that the divergence arises from persistent correlation between the three lowest frequency modes on chains. Sonification of the mode heat fluxes revealed regions where the heat flux amplitude yields a somewhat sinusoidal envelope with a long period similar to the relaxation time. This characteristic in the divergent mode heat fluxes gives rise to the overall thermal conductivity divergence, which strongly supports earlier hypotheses that attribute the divergence to correlated phonon-phonon scattering/interactions. Main text:In all substances, energy/heat is transferred through atomic motions. In electrically conductive materials, electrons can become the primary heat carriers, but in all phases of matter, atomic motions are always present and contribute to the thermal conductivity of every object. Here, we use the term "object" instead of "material" to emphasize that thermal conductivity is highly dependent on the actual structure of an object, that is, its internal atomic level structure and its larger nanoscale or microscale geometry [1][2][3][4][5][6] . In studying the physics of thermal conductivity, tremendous progress has been made over the last 20 years toward understanding various mechanisms that allow one to reduce thermal conductivity in solids 1,7 . This reduction is generally measured relative to a theoretical upper limiting maximum value that arises solely from the intrinsic anharmonicity within the interactions between atoms.In the limiting case, where the atomic interactions are perfectly harmonic, thermal conductivity tends to infinity because the normal modes of vibration do not interact, thus yielding effectively infinite phonon mean free paths and thermal conductivity. As a result, the notion of anharmonicity is critical, and perfectly harmonic interactions between atoms are physically unreasonable. This is because an asymmetry in the energetic well between atoms is an inherent consequence of finite bonding energy (e.g., at some point, the potential energy surface must become concave down). Although one can approach the harmonic limit at cryogenic temperatures, the notion that divergent thermal conductivity (e.g., anomalous thermal conductivity) might be realizable in a real material at room temperature seems theoretically impossible. However, in 1955 Fermi, Pasta, and Ulam (FPU) 8 made a "shocking little discovery" that even an anharmonic system can exhibit infinite thermal conductivity.FPU conducted a numerical experiment with a one-dimensio...

No abstract

Emotion is a word not often heard in sonification, though advances in affective computing make the data type imminent. At times the relationship between emotion and sonification has been contentious due to an implied overlap with music. This paper clarifies the relationship, demonstrating how it can be mutually beneficial. After identifying contexts favourable to auditory display of emotion, and the utility of its development to research in musical emotion, the current state of the field is addressed, reiterating the necessary conditions for sound to qualify as a sonification of emotion. With this framework, strategies for display are presented that use acoustic and structural cues designed to target select auditory-cognitive mechanisms of musical emotion. Two sonifications are then described using these strategies to convey arousal and valence though differing in design methodology: one designed ecologically, the other computationally. Each model is sampled at 15-second intervals at 49 evenly distributed points on the AV space, and evaluated using a publically available tool for computational music emotion recognition. The computational design performed 65 times better in this test, but the ecological design is argued to be more useful for emotional communication. Conscious of these limitations, computational design and evaluation is supported for future development.

Social media is an overwhelmingly visual medium, and we ask the simple question: How can the data and images of social media posts be transformed into something as meaningful and vivid in the auditory sense? Such a design would be useful for eyes-free browsing and could enhance the existing visual media. Our strategy first uses artificial intelligence systems to transform low-level input data into high-level sociocultural features. These features are then conveyed using a multifactored temporal design that uses speech, sonification, auditory scenes, and music.

Informal learning environments (ILEs) like museums incorporate multi-modal displays into their exhibits as a way to engage a wider group of visitors, often relying on tactile, audio, and visual means to accomplish this. Planetariums, however represent one type of ILE where a single, highly visual presentation modality is used to entertain, inform, and engage a large group of users in a passive viewing experience. Recently, auditory displays have been used as a supplement or even an alternative to visual presentation of astronomy concepts, though there has been little evaluation of those displays. Here, we designed an auditory model of the solar system and created a planetarium show, which was later presented at a local science center. Attendees evaluated the performance on helpfulness, interest, pleasantness, understandability, and relatability of the sounds' mappings. Overall, attendees rated the solar system and planetary details very highly, in addition to providing open-ended responses about their entire experience.

The PhET project is a collection of over 130 interactive simulations (or “sims”) designed to teach physics concepts to students from elementary to university levels. The sims rely heavily on visual representation, making them inaccessible to students with disabilities, including those with visual impairments. We present the theory, methods, and process behind our audio design and provide example mapping strategies from two of the simulations. We compare physical, abstract, and musical mapping strategies, noting the strengths of each. We conclude with design recommendations that have arisen in our work, and for which we think would benefit the field at large.