Search

We present a database of parallel recordings of speech and singing, collected and released by the Human Language Technology (HLT) laboratory at the National University of Singapore (NUS), that is called NUS-HLT Speak-Sing (NHSS) database. We release this database to the public to support research activities, that include, but not limited to comparative studies of acoustic attributes of speech and singing signals, cooperative synthesis of speech and singing voices, and speech-to-singing conversion. This database consists of recordings of sung vocals of English pop songs, the spoken counterpart of lyrics of the songs read by the singers in their natural reading manner, and manually prepared utterance-level and word-level annotations. The audio recordings in the NHSS database correspond to 100 songs sung and spoken by 10 singers, resulting in a total of 7 hours of audio data. There are 5 male and 5 female singers, singing and reading the lyrics of 10 songs each. In this paper, we discuss the design methodology of the database, analyse the similarities and dissimilarities in characteristics of speech and singing voices, and provide some strategies to address relationships between these characteristics for converting one to another. We develop benchmark systems, which can be used as reference for speech-to-singing alignment, spectral mapping, and conversion using the NHSS database., Accepted to Speech Communication

We searched for an optimized protocol for mapping observations from a point spectrometer onto a shape model composed of triangular facets, in the context of NASA's asteroid sample return mission, OSIRIS‐REx (Origins, Spectral Interpretation, Resource Identification, and Security‐Regolith Explorer). Our study was conducted before the spacecraft arrived at the mission target asteroid (101955) Bennu, and we used observational sequence plans of the OSIRIS‐REx Visible and InfraRed Spectrometer (OVIRS). We explored six methods of mapping data to shape model facets, using three spatial resolutions. We attempted to boost map fidelity by increasing the observational coverage of the surface. We find that increasing shape model resolution improves mapping quality. However, once the shape model's mean facet edge length is smaller than two‐fifths of the diameter of the instrument's field of view (FOV), the increase in quality tapers off. The six mapping methods can be broken into two categories: facets that (1) select or (2) combine (average) data from observations. The quality differences between similar averaging methods (clipped average, weighted average, etc.) are insignificant. Selecting the nearest observation to a facet best preserves an enclosed outcrop's shape and signal, but averaging spots are more conservative against errors in photometric modeling. A completely enclosed outcrop border expands into the surrounding region by 0.8–1.5 radii of the instrument's FOV. Regions smaller than the instrument's FOV are present in resulting maps; however, their signal strength is reduced as a function of their size relative to the instrument FOV.

This study proposes a multi-microphone complex spectral mapping approach for speech dereverberation on a fixed array geometry. In the proposed approach, a deep neural network (DNN) is trained to predict the real and imaginary (RI) components of direct sound from the stacked reverberant (and noisy) RI components of multiple microphones. We also investigate the integration of multi-microphone complex spectral mapping with beamforming and post-filtering. Experimental results on multi-channel speech dereverberation demonstrate the effectiveness of the proposed approach., to appear in ICASSP 2020

Handwriting verification represents an important area in forensic science. Herein, handwritings based on different types and brands pen ink samples made in China were investigated by non-invasive micro-identification methods. Scanning electron microscopy and energy spectrum (SEM-ES) were employed to measure the micro elements and component analysis of the written traces, and spectral mapping (surface scanning) function was focused. The exact assessments of elements in handwriting traces can distinguish the microcosmic differences between different types and brands ink in handwriting, which provides a new means of analysis and research methods of pen handwriting detection in forensic science.

A mapping technique was used in the present study to explore the biological and imaging characteristics of invasive breast cancer and normal breast tissues in Raman examination data and construct a diagnostic model for breast cancer. Raman examination data reflect the biochemical or molecular characteristics of the target tissues. A total of 45 specimens from patients with breast cancer who underwent surgery and 25 adjacent normal breast tissue specimens were included in the present study. Using the specimens, a total of 53 sets of mapping data and 2,597 pieces of Raman spectral data were obtained. The collected spectra were corrected and fitted, the Raman spectra were analyzed by robust statistical methods, and a diagnostic model was constructed using the k-Nearest Neighbor (KNN) method. The KNN classification method was applied to analyze the characteristics of the mapping test application. The percentage of outliers in the mapping data for malignant and normal breast tissues was 12.7 and 6.6%, respectively. The percentage of outlier data in the conventional single-point detection data for malignant and normal breast tissues was 24.5 and 26.0%, respectively. Analysis using a t-test identified a significant difference in the number of outliers between mapping and single-point detection for malignant (t=-6.169; P

In this paper, we consider the dense correspondence of volumetric images and propose a convolutional network-based descriptor learning framework using the functional map representation. Our main observation is that the correspondence-steered descriptor learning improves dense volumetric mapping compared with the hand-crafted descriptors. We present an unsupervised way to find the optimal network parameters by aligning volumetric probe functions and the enforcement of invertible coupled maps. The proposed framework takes the one-channel volume as input and outputs the multi-channel volumetric descriptors using the cascaded convolutional operators, which are faster than the conventional descriptor computations. We follow the deep functional map framework and represent the dense correspondence by the low-dimensional spectral mapping for the functional transfer and dense correspondence using the linear algebra. We demonstrate that by using the correspondence-steered deep descriptor learning, the quality of both the dense correspondence and attribute transfer are improved in the extensive experiments.

Visual and thermal cameras have large modality gap while providing different information of the same scene. The spectral mapping of thermal imagery into color imagery is a challenging task due to inherit nonlinear relationship. This paper deals with the automatic colorization of thermal imagery into color images using deep encoder-decoder convolutional neural network architecture. The presented approach is trained and evaluated on an online thermal/color dataset, where the network is trained on the spectral mapping of thermal to color images. The training is automatic without requiring user intervention. Extensive experimentation is carried out on the online available dataset, demonstrating consistent and harmonious appearance of generated colorized images from input thermal images.

In this paper we consider the structure of a multifunctional radio receiving center. We talk about the issues of narrowband signals performance and reliability and the spectral mapping storage subsystem. The article presents two methods of reducing the volume of the data recorded and broadcast through the network. We have described the algorithm of the digital signal data packaging and have identified its application limits.

Oral cancer having a high incidence and recurrence rate causes a potential threat and huge economic burden in Southeast Asia. In light of the aforementioned text, early and accurate diagnosis of this disease addressing molecular level changes would play a vital role in reducing the recurrence and increasing the survival rate. Raman spectroscopy has significant potential in the characterization and clinical diagnosis of normal and cancerous tissues at molecular level. To the best of our knowledge, to date, no study has been reported for Raman mapping of oral tissues. This study consistently showed the increased presence of glycogen, lipids, and proteins in all the normal oral tissues, whereas the nucleic acid content was found to be more in cancerous samples. The spectral data were exploited to construct pseudocolor images using principal component analysis and K-means clustering. Thus, obtained Raman maps clearly delineated the tumor margins and was confirmed by the subsequent histopathological report of the hematoxylin–eosin stained samples. A blinded sample was mapped as well, and it showed a few pixels characteristic of malignancy. Upon histological evaluation, the blinded sample was reported as dysplastic, which was in agreement with the obtained Raman image. Hence, this study provides a firm basis to Raman spectral mapping for oral cancer diagnosis and for delineating the tumor margin. Copyright © 2014 John Wiley & Sons, Ltd.

Fusion of the high-spatial-resolution hyperspectral (HHS) image using low-spatial- resolution hyperspectral (LHS) and high-spatial-resolution multispectral (HMS) image is usually formulated as a spatial super-resolution problem of LHS image with the help of an HMS image, and that may result in the loss of detailed structural information. Facing the above problem, the fusion of HMS with LHS image is formulated as a nonlinear spectral mapping from an HMS to HHS image with the help of an LHS image, and a novel cluster-based fusion method using multi-branch BP neural networks (named CF-BPNNs) is proposed, to ensure a more reasonable spectral mapping for each cluster. In the training stage, considering the intrinsic characteristics that the spectra are more similar within each cluster than that between clusters and so do the corresponding spectral mapping, an unsupervised clustering is used to divide the spectra of the down-sampled HMS image (marked as LMS) into several clusters according to spectral correlation. Then, the spectrum-pairs from the clustered LMS image and the corresponding LHS image are used to train multi-branch BP neural networks (BPNNs), to establish the nonlinear spectral mapping for each cluster. In the fusion stage, a supervised clustering is used to group the spectra of HMS image into the clusters determined during the training stage, and the final HHS image is reconstructed from the clustered HMS image using the trained multi-branch BPNNs accordingly. Comparison results with the related state-of-the-art methods demonstrate that our proposed method achieves a better fusion quality both in spatial and spectral domains.

The paper presents a novel artificial bee colony clustering (ABCC) algorithm with a self-adaptive multidimensional search mechanism based on difference bias for insula functional parcellation, called as DABCC. In the new algorithm, the preprocessed functional magnetic resonance imaging (fMRI) data was mapped into a low-dimension space by spectral mapping to reduce its dimension in the initialization. Then, clustering centers in the space were searched by the search procedure composed of employed bee search, onlooker bee search and scout bee search, where a self-adaptive multidimensional search mechanism based on difference bias for employed bee search was developed to improve search capability of ABCC. Finally, the experiments on fMRI data demonstrate that DABCC not only has stronger search ability, but can produce better parcellation structures in terms of functional consistency and regional continuity.

SMI (SPICA Mid-infrared Instrument) is one of the two focal-plane scientific instruments planned for new SPICA, and the Japanese instrument proposed and managed by a university consortium in Japan. SMI covers the wavelength range of 12 to 36 μm, using the following three spectroscopic channels with unprecedentedly high sensitivities: low-resolution spectroscopy (LRS; R = 50 - 120, 17 - 36 μm), mid-resolution spectroscopy (MRS; R = 1300 - 2300, 18 - 36 μm), and high-resolution spectroscopy (HRS; R = 28000, 12 - 18 μm). The key functions of these channels are high-speed dustband mapping with LRS, high-sensitivity multi-purpose spectral mapping with MRS, and high-resolution molecular-gas spectroscopy with HRS. This paper describes the technical concept and scientific capabilities of SMI.

The potential for Raman spectroscopy to provide early and improved diagnosis on a wide range of tissue and biopsy samples in situ is well documented. The standard histopathology diagnostic methods of reviewing H&E and/or immunohistochemical (IHC) stained tissue sections provides valuable clinical information, but requires both logistics (review, analysis and interpretation by an expert) and costly processing and reagents. Vibrational spectroscopy offers a complimentary diagnostic tool providing specific and multiplexed information relating to molecular structure and composition, but is not yet used to a significant extent in a clinical setting. One of the challenges for clinical implementation is that each Raman spectrometer system will have different characteristics and therefore spectra are not readily compatible between systems. This is essential for clinical implementation where classification models are used to compare measured biochemical or tissue spectra against a library training dataset. In this study, we demonstrate the development and validation of a classification model to discriminate between adenocarcinoma (AC) and non-cancerous intraepithelial metaplasia (IM) oesophageal tissue samples, measured on three different Raman instruments across three different locations. Spectra were corrected using system transfer spectral correction algorithms including wavenumber shift (offset) correction, instrument response correction and baseline removal. The results from this study indicate that the combined correction methods do minimize the instrument and sample quality variations within and between the instrument sites. However, more tissue samples of varying pathology states and greater tissue area coverage (per sample) are needed to properly assess the ability of Raman spectroscopy and system transferability algorithms over multiple instrument sites.

In this paper, a supervised feature extraction method, named orthogonal discriminant projection (ODP), is presented. As an extension of spectral mapping method, the proposed algorithm maximizes the weighted difference between the non-local scatter and the local scatter. Moreover, the weights between two nodes of a graph are adjusted according to their class information and local information. Experiments on FERET face data, Yale face data and MNIST handwriting digits data validate that ODP can offer better recognition rate than some other feature extraction methods, such as local preserving projection (LPP), unsupervised discriminant projection (UDP) and orthogonal LPP (OLPP).

The Takab area, located in north-west Iran, is an important gold mineralized region with a long history of gold mining. The gold is associated with toxic metals/metalloids. In this study, Advanced Space Borne Thermal Emission and Reflection Radiometer data are evaluated for mapping gold and base-metal mineralization through alteration mapping. Two different methods are used for argillic and silicic alteration mapping: selective principal-component analysis and matched filter processing (MF). Running a selective principal-component analysis using the main spectral characteristics of key alteration minerals enhanced the altered areas in PC2. MF using spectral library and laboratory spectra of the study area samples gave similar results. However, MF, using the image reference spectra from principal component (PC) images, produced the best results and indicated the advantage of using image spectra rather than library spectra in spectral mapping techniques. It seems that argillic alteration is more effective than silicic alteration for exploration purposes. It is suggested that alteration mapping can also be used to delineate areas contaminated by potentially toxic metals.

We verified the significance of 3-dimensional (3D) spectral mapping during atrial fibrillation (AFIB) using a 64-channel magnetocardiogram (MCG). The study consisted of 16 patients with valvular heart disease who had chronic AFIB. All 16 patients had surgical pulmonary vein (PV) isolation followed by valvular repair. We performed spectral mapping by fast Fourier transform analysis in nonaveraged 64-channel MCG data. The 3D spectral map was superimposed on a 3D heart polygon. At 1 year after surgical PV isolation for AFIB, followed by valve repair, 7 patients had restoration to sinus rhythm, and 9 patients remained in persistent AFIB. The preoperative mean 3D frequency of AFIB was 6.1 ± 0.9 Hz in patients with restored sinus rhythm and 7.2 ± 0.7 Hz in patients with sustained AFIB after PV isolation ( P = .02). In addition, the preoperative 3D spectrum was distributed on the right side of the heart in patients with persistent AFIB. In conclusion, 3D spectral mapping using 64-channel MCG may represent a meaningful noninvasive strategy for patients with AFIB who receive an interventional procedure.

A study of the development of a car-mounted system for mobile hyper-spectral mapping that integrates thermal cameras, near-infrared cameras, and 3D LiDAR data. The data produced by this system is used for city scale thermal energy analysis, which allows property owners to determine the most cost effective energy efficiency improvements for their buildings. The data collection system uses ROS to record several terabytes of data during each night of operation. This case study will consider our internal best practices and lessons learned during development of a robust system running ROS for thousands of miles.

We generate a digital image from spectral mapping by Raman spectroscopy. PCA was applied to obtain representative values of the interest area. The transformed data allowed us to reconstruct an image of the scanning area.

Voice conversion (VC) is the task of modifying a source speaker's voice to match that of a specific target speaker. Traditional methods use Gaussian mixture models (GMM), but the converted speech quality is often badly degraded due to over-smoothing. More recent approaches such as Dynamic Frequency Warping (DFW) maintain more spectrum details during transformation, but require specific formant frequency estimates, with estimation errors resulting in poor similarity between source and target speakers. This paper proposes a new method for voice conversion called Minimum Distance Spectral Mapping (MDSM), based on a frequency-warped point-to-point mapping that robustly and accurately transforms formant frequencies while also maintaining spectral details. The proposed MDSM method uses a minimum distance alignment between source and target speakers, rather than direct formant estimates, which increases robustness and also preserves other spectral details such as formant bandwidth. Results show that the proposed method offers a good trade-off between voice quality and identity similarity, outperforming traditional GMM and DFW in both subjective and objective evaluations.

The effect of the addition of extracts of agro-industrial wastes to the culture media on the production of s-glucosidase, xylanase, laccase, manganese-dependent and independent peroxidases by the edible fungus Pleurotus ostreatus was determined. The relationship between cultivation parameters and the enzyme activities was assessed by spectral mapping technique combined with non-linear mapping. It was proved that extracts enhanced markedly the activities of laccase, manganese-dependent and independent peroxidases. Multivariate mathematical-statistical methods indicated that the enzyme activities were the highest in culture containing pepper extract. It was further demonstrated that the selectivity of the enzyme production was negligible up till 14 days of fermentation and reached the maximum at the 28 th day.

By adding a transmission grating into the optical path of our photon emission system and after calibration, we have completed several failure analysis case studies. In some cases, additional information on the emission sites is provided, as well as understanding of the behavior of transistors that are associated to the fail site. The main application of the setup is used for finding and differentiating easily related emission spots without advance knowledge in light emission mechanisms in integrated circuits.

The MESSENGER spacecraft’s Mercury Atmospheric and Surface Composition Spectrometer (MASCS) obtained more than 1.6 million reflectance spectra of Mercury’s surface from near-ultraviolet to near-infrared wavelengths during the first year of orbital operations. A global analysis of spectra in the wavelength range 300–1450 nm shows little regional variation in absolute reflectance or spectral slopes and a lack of mineralogically diagnostic absorptions. In particular, reflectance spectra show no clear evidence for an absorption band centered near 1 μm that would be associated with the presence of ferrous iron in silicates. There is, however, evidence for an ultraviolet absorption possibly consistent with a very low iron content (2–3 wt% FeO or less) in surface silicates and for the presence of small amounts of metallic iron or other opaque minerals in the form of nano- or micrometer-sized particles. These findings are consistent with MESSENGER X-ray and gamma-ray measurements of Mercury’s surface iron abundance. Although X-ray and gamma-ray observations indicate higher than expected quantities of sulfur on the surface, reflectance spectra show no absorption bands diagnostic of sulfide minerals. Whereas there is strong evidence of water ice in permanently shadowed craters near Mercury’s poles, MASCS spectra provide no evidence for hydroxylated materials near permanently shadowed craters.

Introduction Raman spectroscopy has been shown to accurately classify tissue pathology in a variety of conditions and organ systems. Much of this work has been performed using Raman microspectrometers on tissue sections. Despite the demonstrated potential as an accurate cancer diagnostic tool, Raman spectroscopy (RS) is yet to be adopted by the clinic for histopathology reviews. The Stratified Medicine through Advanced Raman Technologies (SMART) consortium has begun to address some of the hurdles (e.g. tissue sample preparation, data collection, pre-processing and transferability) in its adoption for cancer diagnosis. SMART is a multicentre industry-academic collaboration with the aim of developing a pathology platform for advanced diagnosis, using developments in hardware and software. Renishaw’s Streamline™ Raman technology enables collection of Raman spectral much faster without compromising signal to noise. This study aims to assess the ability of this technique to accurately classify tissue pathology, using an oesophageal model. Methods Specimens were collected from patients with Barrett’s oesophagus (BO), dysplasia and adenocarcinoma, and snap frozen in liquid nitrogen. 8μm tissue sections were prepared onto calcium fluoride slides, with contiguous sections stained with haematoxylin and eosin (H&E) for histological comparison. Raman spectra were collected across homogeneous regions of tissue pathology, using Streamline™ acquisitions of 60 seconds/line, at 1.1μm spatial resolution. Results Advanced multivariate statistical analysis tools were used to develop pathology classification models, which were then tested using leave-one-out cross-validation. Each sample was then classified using a ‘voting classification’ for all pixels from one sample. The sensitivity and specificity of this pathology classification model using Raman Spectroscopy to discriminate dysplasia/adenocarcinoma from Barrett’s oesophagus produced sensitivity and specificities >80%. Conclusion By combining multivariate statistical analysis with Streamline™ Raman acquisition of spectral data, we have demonstrated good sensitivities and specificities. This study illustrates the potential of non-invasive rapid Raman spectral mapping measurements and development of a robust and validated oesophageal classification model that are able to classify tissue pathology. Disclosure of Interest None Declared

One of the most significant problems of detecting communities in networks is the lack of prior knowledge about the number of communities. Here we proposed a correlation spectral mapping based adaptive community detection (CSMACD) algorithm, which employs the concept of the well-known modularity value and the Eigenspectrum of the correlation matrix to automatically detect the network's community structure. Extensive tests on both artificial and real-world networks demonstrate the competitive performance of CSMACD compared to other state-of-the-art community detection algorithms, especially when one has to discover the structure of fuzzy and multi-scale networks.

We present a spectral embedding technique for semi-supervised pattern classification. Importance scores of features are firstly evaluated with a semi-supervised feature selection algorithm by Zhao et al. Training data are then embedded into a low-dimensional space with a spectral mapping derived from the selected and weighted feature vectors with which test data are classified by the nearest neighbor rule. The performance of the proposed pattern classification algorithm is examined with synthetic and real datasets.

Catheter ablation for atrial fibrillation (AF) has emerged as an effective treatment for paroxysmal AF. However, permanent AF ablation success rates are dramatically lower, thus fostering an interest in alternate and adjunctive strategies, such as targeting the complex fractionated atrial electrograms (CFAEs), CFAE in addition to pulmonary vein antrum isolation (PVAI), ablation of AF nests guided by real-time spectral mapping, and ablation of AF nests in addition to PVAI.

FUNDAMENTO: Através de mapeamento espectral-(ME) endocárdico em ritmo sinusal, observam-se dois tipos de miocárdio atrial: o compacto de espectro liso e o fibrilar de espectro segmentado ("Ninho de FA" [NFA]). Durante a FA o compacto tem ativação organizada e baixa freqüência (passivo) enquanto o fibrilar apresenta ativação bastante desorganizada e alta freqüência (ativo/ressonante) sendo ambos ativados por uma taquicardia protegida de alta freqüência, taquicardia de background (TB). OBJETIVO: Descrever o tratamento da FA pela ablação dos NFA e da TB. MÉTODOS: 1) Ablação por cateter-RF [4/8mm-60º/30-40J/30s] dos NFA guiada por ME em ritmo sinusal, fora das veias pulmonares; 2) Estimulação atrial-300ppm; 3) Ablação adicional de NFA se induzida FA; 4) Ablação focal se induzida TB e/ou Flutter; 5) Seguimento clínico+ECG+Holter. RESULTADOS: Foram tratados 50±18 NFA/paciente. Após 11,3±8m 81p (88%) estavam sem FA (28,3% com antiarrítmico). Após a ablação dos NFA: a FA não foi reinduzida em 61p(71%); TB foi induzida e tratada em 24p(26%). Ocorreram 2 sangramentos pericárdicos (1 tratado clinicamente e 1 cirurgicamente) ocasionados por bainhas não mais utilizadas. CONCLUSÃO: O ME em ritmo sinusal ablaciona os NFA. Durante a FA os NFA apresentam um padrão reativo-ressonante e o miocárdio compacto apresenta-se passivo, estimulados pela alta freqüência da TB. Após a ablação dos NFA e da TB não foi possível reinduzir FA sustentada. A ablação dos NFA fora das VP se mostrou segura e altamente eficiente para a cura e/ou o controle clínico da FA. BACKGROUND: Two types of myocardia can be observed through the endocardial spectral mapping (SM) in sinus rhythm: the compact type with a smooth spectrum and the fibrillar type with a segmented spectrum (atrial fibrillation nests). During the atrial fibrillation (AF), the compact type has an organized activation and low frequency (passive), whereas the fibrillar type has a rather disorganized activation and high frequency (active/resonant), with both being activated by high-frequency sustained tachycardia - the background tachycardia (BT). OBJECTIVE: To describe the treatment of AF by the ablation of the AF nests and BT. METHODS: 1) Catheter ablation of the AF nests with RF [4/8mm-60º/30-40J/30s] guided by SM in sinus rhythm, outside the pulmonary vein; 2) atrial stimulation -300ppm; 3) Additional ablation of the AF nests if AF is induced; 4) Focal ablation if BT and/or Flutter is induced; 5)Clinical follow-up+ ECG+ Holter. RESULTS: A total of 50±18 AF nests/patient were treated. After 11.3±8m, 81 patients (88%) did not present AF (28.3% with antiarrhythmic drugs). After the ablation of the AF nests, AF was not reinduced in 61 patients (71%) and BT was induced and treated in 24 patients (26%). There were two episodes of pericardial bleeding (1 treated clinically and 1 surgically), caused by sheaths that are no longer used CONCLUSION: The SM in sinus rhythm can be used in the ablation of AF nests. During the AF, the AF nests present a reactive-resonant pattern and the compact myocardium is passive, stimulated by the high frequency of the BT. After the ablation of the AF nests and the BT, it was not possible to reinduce the sustained AF. The Ablation of AF nests outside the pulmonary veins showed to be safe and highly effective in the cure and/or clinical control of the AF.

Introduction Raman spectroscopy has been shown to accurately classify tissue pathology in a variety of conditions and organ systems. Much of this work has been performed using Raman microspectrometers on ex vivo tissue sections with long acquisition times, and measurements can take many hours. To enable translation of this technology to a clinical setting, either as an adjunct for pathologists or an in vivo probe for point of care testing, measurement times must be reduced. By using Renishaw’s Streamline™ technologywhich combines a line focusing technique and pixel binning, it is possible to collect Raman spectral measurements much faster without compromising signal to noise. This study aims to assess the ability of this technique to accurately classify tissue pathology, using an oesophageal model. Method Tissue was collected from the oesophagus in patients undergoing endoscopy of open surgery on the oesophagus. Specimens were collected from patients with Barrett’s oesophagus (BO), dysplasia and adenocarcinoma, and snap frozen in liquid nitrogen. 8 µm tissue sections were prepared onto calcium fluoride slides, with contiguous sections stained with haematoxylin and eosin (H&E) for histological comparison. Raman spectra were collected across homogeneous regions of tissue pathology, using Streamline™ acquisitions of 60 s per pixel, at 1.1 µm spatial resolution. Classification models were constructed to discriminate pathology subtypes. Results Advanced multivariate statistic analysis tools were used to develop pathology classification models, which were then tested using leave-one-out cross-validation. The sensitivity and specificity of this pathology classification model using Raman Spectroscopy to discriminate dysplasia/adenocarcinoma from Barrett’s oesophagus produced sensitivity and specificities >75% and >75%, respectively. Conclusion By combining multivariate statistical analysis with Streamline TM Raman acquisition of spectral data, we have demonstrated good sensitivities and specificities. This study illustrates the potential of non-invasive rapid Raman spectral mapping measurements and development of a robust and validated oesophageal classification model that are able to classify tissue pathology both providing a clinical tool for pathologists and clinicians. Disclosure of interest None Declared. Reference Shetty G, et al . Raman spectroscopy: elucidation of biochemical changes in carcinogenesis of oesophagus. Br J Cancer. 2006;94(10):1460–4

The fundamentals of radiation calculations in planetary atmospheres have been known for a long time, but the practical problems of calculating radiation over a broad spectral range both efficiently and accurately remain an area of active research. The Goody et al. paper of 1989 [13] established a milestone in that long effort. Goody et al. highlighted attempts to formulate the problem in terms of a few absorption coefficients that could represent typically tens of thousands of coefficients in a rigorous line-by-line calculation. Here we provide a brief background, point out that the correlated-k method is a special case of a broader spectral mapping concept, and mention some new ideas that have emerged recently.

We demonstrate the ability of a combined scanning electron microscope and cathodoluminescence (CL) spectral mapping system to provide important spatially resolved information. The degree of inhomogeneity in spectral output across a multi-quantum well sample is measured using the SEM-CL system as well as measuring the efficiency roll-off with increasing carrier concentration. The effects of low energy electron beam modification on the InGaN/GaN multi quantum wells have also been characterized.

The atrial cycle length during atrial fibrillation (AFIB) and flutter (AFL) waves has been shown to be correlated with atrial refractoriness and organization. In this study, a novel body surface 2-dimensional (2D) spectral mapping of AFIB and AFL was developed using vector-projected 187-channel ECG (187ch DREAM-ECG). The study consisted of 28 patients (24 AFIB and 4 AFL with valvular heart disease). Spectral analysis of a 4 seconds of non-averaged 187ch wave during AFIB and AFL was performed by maximum entropy modeling. The 2D body surface spectral feature was displayed according to the frequency and power magnitude components of the dominant frequency. The accuracy of this spectral feature was verified by 64ch magnetocardiography (MCG). The average dominant frequency in AFL by 187ch VP-ECG was lower than that in AFIB (4.6±0.9 Hz in AFL vs. 7.2±0.9 Hz in AFIB). The average dominant frequency by 187ch SACEG was agreed well with those by 64ch MCG (y=0.86x+0.84, r2=0.89). The body surface 2D spectral feature of the dominant frequency demonstrated a homogenous spectrum pattern in AFL and a non-homogenous pattern in AFIB. In conclusion, 2D novel body surface spectral mapping of the dominant frequency using 187ch DREAM-ECG may evaluate non-invasively the atrial cycle length during AFIB and AFL.

Spectral mapping is a classification technique that has been applied to the Liege physiognomies of the neuroleptics. The method makes use of a special projection that separate potency from spectral information. Principal component analysis revealed a dominant component resembling a bipolar incisive/sedative scale. A second antimanic/antiautistic scale is apparent in the revised. Liege physiognomie. Spectral mapping allows to compare observations on neuroleptics in different frames of reference and from different methodologies (e.g. clinical and pharmacological).

In this work we investigate the use of a speaker adaptation technique, for speech recognition, based on neural network spectral mapping. Different multilayer perceptron neural network architectures are analyzed in order to optimize the spectral difference reduction in acoustic data of two speakers. Experiments are carried out in a telecontrol environment, used to provide voice commands to a mobile robot, based on DTW pattern matching.

This paper is the first reported description of Raman Spectroscopy in the assessment of axillary lymph nodes in breast cancer patients. Raman Spectroscopy is an inelastic scattering spectroscopic technique appropriate for the assessment of unprocessed complex biological tissues. Spectra represent biochemical signatures of the tissue under scrutiny. The described method of Raman spectral mapping produces false-color spectral images of lymph node sections. These can be compared with standard histopathology slides and white light images of nodal tissue. This method has the potential to allow the detailed biochemical assessment of heterogeneous lymph node features, and to contribute towards the development of an optical diagnostic tool for use in a clinical setting.

A variant of electroencephalogram (EEG) power spectral mapping called power ratio index (PRI) mapping was used to monitor 15 patients with malignant brain tumors. This index is generated by dividing the low frequency (delta, theta) power by the high frequency (alpha, beta) power. Because the nonparoxysmal effect of a brain tumor on the EEG is reflected as a relative loss of high frequency power and a gain in low frequency power, utilization of the PRI has the effect of placing the epicenter of the “power dysfunction” coincident with the epicenter of the tumor.