E. Andrade Lima

Magnetic Measurement Techniques for Locating Foreign Bodies in Humans

There is a high incidence of needles accidentally introduced in the human body. The commonly used radiological techniques present several limitations. They usually do not give the appropriate informations for a surgical extraction procedure, like the depth relative to the skin and only the neddle projections are available. The development of a more accurate, non-invasive and innocuous method for a magnetic foreign body localization is of meaningful clinical importance. In this way, non-invasive magnetic field measurements, using a SQUID magnetometer, were performed to locate, for surgical purposes, needles lost in the bodies of 5 children. Also, another approach for detecting magnetic fields generated by foreign bodies, using a fluxgate magnetometer, is proposed.

A survey of calibration procedures for SQUID gradiometers

The accuracy of three distinct experimental procedures for calibrating axial SQUID gradiometers has been compared, for the same gradiometer design and experimental setup. Each procedure employs a different field source, and a nonlinear least-squares optimization is used to fit the measured voltage to the theoretical field, thus determining Tesla/Volt calibration factors. We also studied the effect of noise and gradiometer imbalance on the accuracy of each procedure.

Flux/voltage calibration of axial SQUID gradiometers using an optimization procedure

A straightforward experimental procedure for calibration of axial SQUID gradiometers has been developed, based on numerical optimization techniques. A dc current carrying wire of finite length, whose magnetic field spatial distribution is well known, was scanned by a SQUID system at several lift-off distances. Initially, theoretical magnetic field parameters such as lift-off and scanning tilt angles were numerically optimized in order to match the normalized shapes of experimental and theoretical signals. After that, the calibration factor can be easily found as the ratio between the two non-normalized signals. Once the calibration factor was obtained, an experimental validation was made by using a current-carrying copper sheet and by comparing the calibrated experimental signal with a model prediction, leading to good results. The overall procedure is easily implemented and can be modified to account for different SQUID systems and gradiometer geometry.

Image reconstruction of spherical inclusions in ferromagnetic structures using the generalized inverse

The generalized inverse method and singular value decomposition were used to reconstruct the magnetic image of small spherical inclusions in ferromagnetic structures. These techniques were used on the magnetic flux leakage generated when the structure is magnetized. The leakage field from the inclusions was simulated by using a simple constant-permeability model. The recovery model used a grid of magnetic dipoles at the region of the inclusions. The ability of the technique to distinguish between different distributions of inclusions with similar leakage fields is analyzed. The results demonstrates the feasibility of imaging multiple inclusions within the structure.

Image Processing Techniques to Remove Depth Bias Effects in Magnetic Source Images of Deep Cracks

Magnetic Flux Leakage (MFL) methods are used to detect localized phenomena such as surface or sub-surface cracks in ferromagnetic materials [1]. A dc magnetic field is induced inside the sample being tested, and the distribution of the resultant lines of magnetic flux is determined by the values of magnetic permeability within the region of interest. Characteristically, the magnetic flux“leaks” out of the object in the region of a defect, allowing it to be detected using some kind of magnetic sensor.

Application of a single-channel SQUID magnetometer for non-invasive study of cardiac tachyarrhythmias mechanisms

Abstract The underlying electrical mechanism associated with self-sustained arrhythmias such as cardiac flutter and fibrillation is not completely elucidated. Most of the evidence points towards a reentry path, but the hypothesis of firing of an ectopic focus with high frequency is still being considered. This paper evaluates the possibility of distinguishing between reentrant movement and focal excitation pulse propagation through non-invasive magnetic measurements using a SQUID system. We present experiments performed on rabbit atrial tissues immersed in a nourishing solution and submitted to different propagation patterns. The magnetic measurements were made inside a shielded chamber, by a single-channel low- T c rf-SQUID magnetometer coupled to a second-order axial gradiometer with 1.5 cm diameter coils and 4 cm baseline. The magnetic signals have been processed to yield amplitude maps, which show distinct behavior for the two different propagation mechanisms. The results obtained agree with the expected magnetic field behavior according to previous simulated studies based on cellular automata models. Therefore, the potential of the non-invasive magnetocardiographic technique for distinguishing between the primary possibilities of propagation mechanisms is corroborated, with implications in electrophysiology and clinical diagnostic studies.

Application of a MLP neural network for compensation of distortions in strain gauges measurements due to temperature variations in offshore operation of a flexible pipe-lay vessel

Strain gauges are often affected by temperature variations, yielding unacceptable distortions on load measurements. Such effects have been observed on a flexible pipe-lay vessel, in which the load indication presents a large distortion around noon everyday. In this paper, a neural network scheme is employed to overcome such difficulties.

Induced Atrial Flutter in Rabbit Hearts Tissue and Inverse Magnetocardiography

The possibility of non-invasive magnetic identification of electrical tissue activation during atrial flutter in humans is of great clinical interest. Most of the known evidence through electrical [1,2,3] and magnetic [4,5] measurements suggests the existence of a circular motion of the electrical activation (reentry) pattern as the underlying mechanism of this arrhythmia. A new animal experimentation protocol has been developed here in order to allow controlled studies of the reentiy circuit in isolated pieces of rabbit atrium during experimentally induced flutter arrhythmia. In this preparation, the size of the tissue, its anatomical characteristics and the distance from the plane of the sample to the magnetic sensor can be taken into account. The underlying basis for the present work is the “equivalent circular dipole model” presented elsewhere [4]. Therefore, both the animal experimentation and signal processing protocols were based on the rotationally invariable structure of this model. Such stnicture led to good results, overcoming experimental difficulties due to small number of measurement positions and low sensor sensitivity.

Imaging Flaws under Insulation Using a Squid Magnetometer

Superconducting QUantum Interference Devices (SQUID) are the most sensitive instruments known for the measurement of magnetic fields. An all niobium two-hole homemade SQUID can easily achieve sensitivities of 10-4 Ф0/√Hz (Ф0 = 2.07 × 10-15 Wb). Our complete system has a sensitivity of 50 × 10-15 Tesla √Hz, and more sophisticated systems can reach sensitivities one order of magnitude higher. Due to its high sensitivity, and to the advent of high temperature superconductivity, SQUID systems presents new opportunities for its use in nondestructive evaluation of electrically conducting and ferromagnetic structures, mainly when the area to be inspected is difficult to be reached.