P. Marche

A new robotic mechanism for medical application

The difficulty in interpreting ultrasound images and the scarcity of trained ultrasonographers to manipulate a transducer to the optimal position are the basic premises for establishing the SYRTECH project (Tele-Scanning Robot System). The principle objective of this project is to enable an expert ultrasonographer to perform an ultrasound scan examination on a remote patient using master and slave stations. The master is the doctors post and the slave is the robotic system supporting the scan probe. The clinical expert is then able to interpret the transmitted images and move the slave arm in order to obtain the best orientation of the ultrasound beam on any anatomical part of the remotely located patient. This paper presents the master/slave system developed in our laboratory and the chosen robotic solution.

Wavelet-Based Speckle Noise Reduction in Ultrasound B-Scan Images

Speckle noise is known to be signal-dependent in ultrasound imaging. Hence, separating noise from signal becomes a difficult task. This paper describes a wavelet-based method for reducing speckle noise. We derive from the model of the displayed ultrasound image the optimal wavelet-domain filter, in the least mean-square sense. Simulations on synthetic data have been carried out in order to assess the performance of the proposed filter with regards to the classical wavelet shrinkage scheme, while phantom and tissue images have been used for testing it on real data. The results show that the filter effectively reduces the speckle noise while preserving resolvable details.

Automated neural network detection of wavelet preprocessed electrocardiogram late potentials

The aim of the study is to investigate the potential of a feedforward neural network for detecting wavelet preprocessed late potentials. The terminal parts of a simulated QRS complex are processed with a continuous wavelet transform, which leads to a time-frequency represenation of the QRS complex. Then, diagnostic feature vectors are obtained by subdividing the representations into several regions and by processing the sum of the decomposition coefficients belonging to each region. The neural network is trained with these feature vectors. Simulated ECGs with varying signalto-noise ratios are used to train and test the classifier. Results show that correct classification ranges from 79% (high-level noise) to 99% (no noise). The study shows the potential of neural networks for the classification of late potentials that have been preprocessed by a wavelet transform. However, clinical use of this method still requires further investigation.

In vitro 3D reconstruction of long bones using B-scan image processing

An echographic image processing method has been developed, and validated by in vitro experiments, for the 3D reconstruction of the long bones of the newborn. The reconstruction of successive parallel cross-sections is obtained by a 2D reconstruction technique using radial B-scan image processing. The automatic segmentation of all the calculated images allows the extraction of the external contours of the skeleton. After structuring the explored volume using a contour association method, a contour interpolation step is required to solve the anisotropy problem, to obtain a 3D representation with cubic voxel lists. The results are encouraging, and a new mechanical part prototype of the acquisition system is under test for in vivo experiments. The main originality of the paper lies in the combination of different steps to obtain a practical solution to a clinical problem.

Atmospheric trace species measured above haute-provence observatory

Abstract Trace species have been measured in June 1981, at the “Observatoire de Haute-Provence” Chiran station, 1905 m high, with a ground based high resolution i.r. spectrometer, using the Sun as a source. Column abundance amounts are given for methane, formaldehyde, hydrogen chloride and hydrogen fluoride.

Total atmospheric ozone measured by ground based high resolution infrared spectra-comparison with dobson measurements

Abstract Total atmospheric ozone has been measured at the “Observatoire de Haute-Provence”, Chiran Station, 1905 m high with a ground based, high resolution i.r. spectrometer, using the Sun as a source. Results are compared with a Dobson spectrophotometer which has been calibrated and intercompared with a standard. Total ozone values found with Dobson are 4% lower than values given by i.r. measurements. In addition, a negative correlation between total ozone and tropopause height has been found.

Quick primitives extraction using inertia matrix on measures issue from an ultrasonic network

An autonomous mobile robot needs externals sensors for its own localization and modeling environment. In order to perform the angular precision of ultrasonic sensor and to improve the number of informations for a given time, we developed an ultrasonic sensor network using the multiple receivers mode. In this paper we introduce the ultrasonic network with its hardware architecture and its functioning mode. We also present a method to localize an echo source with the multiple receivers mode and show how to simply model this echo in the robot frame. We then describe the extraction of environment primitives (segment) by the inertia matrix and Hough transform. Finally, we validate the all process by experiments.

Wavelet-based enhancement of lesion detectability in ultrasound B-scan images

A new technique for enhancing lesion detectability in ultrasound B-scan images is presented. A wavelet based denoising algorithm is proposed. After discussing the mathematical model of the B-scan images, the authors derive the optimal wavelet shrinkage scheme for the model. As such a filter needs exact knowledge of the noise-free signal and the noise statistics, a classical wavelet shrinkage is used for the approximation of the noise-free signal. The performance of this algorithm is compared to the Maximum Likelihood estimator of the B-scan image by using as a figure of merit the area under the ROC curve. The results have shown that the present algorithm is superior in lesion detection, and an improvement in the area under ROC curve of the order of 0.1 is observed.

Measurements of total ozone: Intercomparison of data from a variety of instruments, June 1981

Abstract The total ozone data obtained over southern France from a variety of instruments during the Intercomparison Campaign of June 1981 are presented and intercompared. All the instruments intercompared produced a self-consistent data set in terms of the variations that occurred during the intercomparison period and a discussion of the differences in the absolute values measured is given.

Tropospheric, stratospheric and mesospheric ozone observed using the Umkehr method

Abstract Ozone vertical profiles have been obtained above the “Observatoire de Haute-Provence” from 9 to 19 June 1981, with both conventional and short Umkehr methods: the mesospheric and stratospheric ozone variability has been found to be very small while the ozone variability was large in the troposphere and maximum at the tropopause level.