A comparison between randomly alternating imaging, normal laparoscopic imaging, and virtual reality training in laparoscopic psychomotor skill acquisition

Objectives: To evaluate virtual reality as a laparoscopic training device in helping surgeons to automate to the “fulcrum effect” by comparing it to time-matched training programs using randomly alternating images (ie, y-axis inverted and normal laparoscopic) and normal laparoscopic viewing conditions.

Methods: Twenty-four participants (16 females and 8 males), were randomly assigned to minimally invasive surgery virtual reality (MIST VR), randomly alternating (between y-axis inverted and normal laparoscopic images), and normal laparoscopic imaging condition. Participants were requested to perform a 2-minute laparoscopic cutting task before and after training.

Results: In the test trial participants who trained on the MIST VR performed significantly better than those in the normal laparoscopic and randomly alternating imaging conditions.

Conclusion: The results show that virtual reality training may provide faster skill acquisition with particular reference to automation of the fulcrum effect. MIST VR provides a new way of training laparoscopic psychomotor surgical skills.

A machine vision approach to the grading of crushed aggregate

The grading of crushed aggregate is carried out usually by sieving. We describe a new image-based approach to the automatic grading of such materials. The operational problem addressed is where the camera is located directly over a conveyor belt. Our approach characterizes the information content of each image, taking into account relative variation in the pixel data, and resolution scale. In feature space, we find very good class separation using a multidimensional linear classifier. The innovation in this work includes (i) introducing an effective image-based approach into this application area, and (ii) our supervised classification using wavelet entropy-based features.

Virtual state scattering with cold electrons: para-xylene and para-difluorobenzene

The scattering of electrons with kinetic energies down to a few meV by para-xylene and para-difluorobenzene has been observed experimentally with an electron beam energy resolution of 0.95 to 1.5 meV (full width half maximum). At low electron energies the collisions can be considered as cold scattering events because the de Broglie wavelength of the electron is considerably larger than the target dimensions. The scattering cross sections measured rise rapidly at low energy due to virtual state scattering. The nature of this scattering process is discussed using s- and p-wave phase shifts derived from the experimental data. Scattering lengths are derived of, respectively, -9.5+/-0.5 and -8.0+/-0.5 a.u. for para-xylene and para-difluorobenzene. The virtual state effect is interpreted in terms of nuclear diabatic and partially adiabatic models, involving the electronic and vibronic symmetries of the unoccupied orbitals in the target species. The concept of direct and indirect virtual state scattering is introduced, through which the present species, in common with carbon dioxide and benzene, scatter through an indirect virtual state process, whereas other species, such as perfluorobenzene, scatter through a direct process. (C) 2005 American Institute of Physics.