A noninvasive eye tracking system based on infrared 3-D video-oculographic techniques is proposed for the automatic monitoring of eye position and orientation in external beam radiotherapy of ocular tumors. The presented method can be applied for the real-time estimation of lesion position and tumor-beam misalignments, allowing automatic patient setup and eye movement gated treatments. A prototypal eye tracker was developed and tested on five subjects, achieving gaze estimation errors of 0.5° and eye monitoring frequencies of 125 Hz. The proposed application can potentially improve quality and efficacy of ocular radiotherapy treatments, currently based on invasive, qualitative, and manual control procedures.

In this work three ECG derived respiration (EDR) methods based on multi leads ECG are applied to a single lead, to evaluate the capabilities to use this method with wearable devices. Preliminary result shows the R-R interval method applied to two wire single lead ECG is more suitable to extract the fundamental respiratory frequency.

Three-dimensional reconstruction of human diaphragm provide useful information on the functional anatomy of the respiratory system by the analysis of its geometry. The aim of the current work is the development of a new method for the 3D analysis of diaphragm geometry by the combination of free-hand Ultrasound (US) scans with an optoelectronic system of movement analysis for the tracking of the probe. 6 healthy subjects (age=24±2), have been measured with a free-hand US scanning of the abdomen in supine (SP) and seated position (ST), during breath-holding at Functional Residual Capacity (FRC) and Total Lung Capacity (TLC). For one of them, measurements have been repeated in supine position in an MR scan for the validation of the method. Posture has different implications on diaphragm geometry depending on the respiratory volume. At FRC the ray of curvature (ρ) in ST is 101.1±43.1 mm higher than in SP (p=0.006), while at TLC, posture influences the position of the diaphragm with a caudo-cranial displacement from SP to ST of 23.4±16.7mm (p=0.019). Diaphragm geometry at different lung volumes is influenced by the posture. In SP, ρ increases of 105.9±48.3 mm (p=0.008) and there is a cranio-caudal displacement (Δy) of 47.54±15 mm (p=0.002) shifting from FRC to TLC. In ST, Δy=31.1±13.5 mm (p=0.006) while ρ increases not significantly. Percentage errors between MR and US 3D reconstructions are 2.1% and 10.46% for Δy and TLC/FRC ρ ratio, respectively. US 3D reconstruction is a reliable method for the assessment of diaphragm functional anatomy. Posture directly influence diaphragm geometry and hence respiratory mechanics.

Calibrated ultrasound (US) probes, formerly used in neurosurgery for acquiring bone implanted markers, have been recently introduced to perform non-invasive registration during computer assisted orthopaedic surgery total knee replacement. In this study an experimental set-up system to detects the surface of the bone using a 3D US probe navigation procedures non-invasively is presented. Preliminary test shows a reconstruction error of about 0.6 +/- 0.39mm.

Three-dimensional reconstruction of human diaphragm provides useful information on the functional anatomy of the respiratory system by the analysis of its geometry. Up to now, MR and CT scans enabled 3D reconstructions of the diaphragm with some limitations: possibility to obtain data only in supine postures, exposition of the subject to ionizing radiations, high costs. Traditional 2D Ultrasound (US) scans allow the analysis of the diaphragm in different postures but do not give any information about its shape. Thus, the aims of the current study were to develop a new method for the analysis of diaphragm 3D geometry by the combination of free-hand US scans and by optoelectronic tracking of the probe and to study diaphragm 3D geometry at different lung volumes and in different postures: supine and seated.

The aim of this paper is to assess the accuracy and the repeatability of measures made by means of a stereophotogrammetric equipment and processed by a special software developed within the EUROShoE project (Growth Programme, European Commission), with the purpose to collect three-dimensional data from the surface of the foot. A prosthetic foot was marked with black stickers (5 mm diameter) corresponding to 10 anatomical landmarks, and repeated acquisitions of its surface were performed within a day and through days. Measures of the distance between pairs of markers were made manually on the artificial foot and estimated through the acquired surface data. The mean accuracy computed on 14 distances was 1.1 mm (SD = 1.2). The intra and inter day variability was assessed by considering repeated measures of conventional shoe making variables (e.g. stick length, ball girth, instep girth,...) produced by using dedicated software. The results, expressed in term of coefficients of variation, ranged between 0.2 and 0.8%. The achieved results demonstrate that the equipment and the software ensure adequate measures for the shoe making process (i.e. characterization and choice of the last) and for an easy detection of the human foot surface. Both of them are key steps for the implementation of best fit or custom made shoe production, even if attention must be paid to enviromental settings during data collection in a retail shop contest.

A "managed care" model (prevention, diagnosis, therapy, admission, home assistance) has been realized to permit both a continuous patients check without hospitalization and a patients data archiving. An hardware and software architecture has been designed to manage the out-of-hospital treatments of patients with a moderate cardiovascular risk; as all the sensors built are wearable the cardio respiratory recording may have place during patients exercise program. Transmitted by wireless ECG has been automatically processed by "Windows Media Center" through a multi parametric approach based on time and frequency (linear and non linear analysis) domain study. Advanced characteristics have been utilized by this monitoring prototype and different technologies, such as data acquisition, elaboration, on line transmission and web-based data storage, have been integrated for knowledge management, in this way it is possible not only data real time visualization and collection but also a continuous patients management useful to improve life quality.

Global polynomial (GP) methods have been widely used to correct geometric image distortion of small-size (up to 30 cm) X-ray image intensifiers (XRIIs). This work confirms that this kind of approach is suitable for 40 cm XRIIs (now increasingly used). Nonetheless, two local methods, namely 3rd-order local un-warping polynomials (LUPs) and hierarchical radial basis function (HRBF) networks are proposed as alternative solutions. Extensive experimental tests were carried out to compare these methods with classical low-order local polynomial and GP techniques, in terms of residual error (RMSE) measured at points not used for parameter estimation. Simulations showed that the LUP and HRBF methods had accuracies comparable with that attained using GP methods. In detail, the LUP method (0.353 μm) performed worse than HRBF (0.348 μm) only for small grid spacing (15×15 control points); the accuracy of both HRBF (0.157 μm) and LUP (0.160 μm) methods was little affected by local distortions (30×30 control points); weak local distortions made the GP method poorer (0.320 μm). Tests on real data showed that LUP and HRBF had accuracies comparable with that of GP for both 30 cm (GP: 0.238 μm; LUP: 0.240 μm; HRBF: 0.238 μm) and 40 cm (GP: 0.164 μm; LUP: 0.164 μm; HRBF: 0.164 μm) XRIIs. The LUP-based distortion correction was implemented in real time for image correction in digital tomography applications.

In this paper we present two novel techniques, namely a local unwarping polynomial (LUP) and a hierarchical radial basis function (HRBF) network, to correct geometric distortions in XRII images. The two techniques have been implemented and compared, in terms of residual error measured at control and intermediate points, with local and global methods reported in the previous literature. In particular, LUP rests on a locally optimized 3rd degree polynomial applied within each quadrilateral cell on the rectilinear calibration grid of points. HRBF, based on a feed-forward neural network paradigm, is constituted by a set of hierarchical layers at increasing cut-off frequency, each characterized by a set of Gaussian functions. Extensive experiments have been performed both on simulated and real data. In simulation, we tested the effect of pincushion, sigmoidal and local distortions, along with the number of calibration points. Provided that a sufficient number of cells of the calibration grid is available, the obtained accuracy for both LUP and HRBF is comparable to or better than that of global polynomial technique. Tests on real data, carried out by using two different (12 in. and 16 in.) XRIIs, showed that the global polynomial accuracy (0.16+/-0.08 pixels) is slightly worse than that of LUP (0.07+/-0.05 pixels) and HRBF (0.08+/-0.04 pixels). The effects of the discontinuity at the border of the local areas and the decreased accuracy at intermediate points, typical of local techniques, have been proved to be smoothed for both LUP and HRBF.

An x-ray image intensifier (XRII) has many applications in diagnostic imaging, especially in real time. Unfortunately the inherent and external distortions (pincushion, S-distortion and local distortion) hinder any quantitative analysis of an image. In this paper an automatic real-time local distortion correction method for improving the quality of digital linear tomography images is presented. Using a digital signal processing (DSP), this method can work for an image up to 1K x 1K x 12 bit at 30fps. A local correction method has been used because it allows distortions such as those caused by poor axial alignment between the X-Ray cone beam and the XRII input surface and local distortions to be resolved that are generally neglected by global methods.