The analytical anisotropic algorithm (AAA) was implemented in the Eclipse (Varian Medical Systems) treatment planning system to replace the single pencil beam (SPB) algorithm for the calculation of dose distributions for photon beams. AAA was developed to improve the dose calculation accuracy, especially in heterogeneous media. The total dose deposition is calculated as the superposition of the dose deposited by two photon sources (primary and secondary) and by an electron contamination source. The photon dose is calculated as a three-dimensional convolution of Monte-Carlo precalculated scatter kernels, scaled according to the electron density matrix. For the configuration of AAA, an optimization algorithm determines the parameters characterizing the multiple source model by optimizing the agreement between the calculated and measured depth dose curves and profiles for the basic beam data. We have combined the acceptance tests obtained in three different departments for 6, 15, and 18 MV photon beams. The accuracy of AAA was tested for different field sizes (symmetric and asymmetric) for open fields, wedged fields, and static and dynamic multileaf collimation fields. Depth dose behavior at different source-to-phantom distances was investigated. Measurements were performed on homogeneous, water equivalent phantoms, on simple phantoms containing cork inhomogeneities, and on the thorax of an anthropomorphic phantom. Comparisons were made among measurements, AAA, and SPB calculations. The optimization procedure for the configuration of the algorithm was successful in reproducing the basic beam data with an overall accuracy of 3%, 1 mm in the build-up region, and 1%, 1 mm elsewhere. Testing of the algorithm in more clinical setups showed comparable results for depth dose curves, profiles, and monitor units of symmetric open and wedged beams below dmax. The electron contamination model was found to be suboptimal to model the dose around dmax, especially for physical wedges at smaller source to phantom distances. For the asymmetric field verification, absolute dose difference of up to 4% were observed for the most extreme asymmetries. Compared to the SPB, the penumbra modeling is considerably improved (1%, 1 mm). At the interface between solid water and cork, profiles show a better agreement with AAA. Depth dose curves in the cork are substantially better with AAA than with SPB. Improvements are more pronounced for 18 MV than for 6 MV. Point dose measurements in the thoracic phantom are mostly within 5%. In general, we can conclude that, compared to SPB, AAA improves the accuracy of dose calculations. Particular progress was made with respect to the penumbra and low dose regions. In heterogeneous materials, improvements are substantial and more pronounced for high (18 MV) than for low (6 MV) energies.

In this work, a novel three-dimensional superposition algorithm for photon dose calculation is presented. The dose calculation is performed as a superposition of pencil beams, which are modified based on tissue electron densities. The pencil beams have been derived from Monte Carlo simulations, and are separated into lateral and depth-directed components. The lateral component is modeled using exponential functions, which allows accurate modeling of lateral scatter in heterogeneous tissues. The depth-directed component represents the total energy deposited on each plane, which is spread out using the lateral scatter functions. Finally, convolution in the depth direction is applied to account for tissue interface effects. The method can be used with the previously introduced multiple-source model for clinical settings. The method was compared against Monte Carlo simulations in several phantoms including lung- and bone-type heterogeneities. Comparisons were made for several field sizes for 6 and 18 MV energies. The deviations were generally within (2%, 2 mm) of the field central axis d(max). Significantly larger deviations (up to 8%) were found only for the smallest field in the lung slab phantom for 18 MV. The presented method was found to be accurate in a wide range of conditions making it suitable for clinical planning purposes.

The timing of individual neuronal spikes is essential for biological brains to make fast responses to sensory stimuli. However, conventional artificial neural networks lack the intrinsic temporal coding ability present in biological networks. We propose a spiking neural network model that encodes information in the relative timing of individual spikes. In classification tasks, the output of the network is indicated by the first neuron to spike in the output layer. This temporal coding scheme allows the supervised training of the network with backpropagation, using locally exact derivatives of the postsynaptic spike times with respect to presynaptic spike times. The network operates using a biologically-plausible alpha synaptic transfer function. Additionally, we use trainable synchronisation pulses that provide bias, add flexibility during training and exploit the decay part of the alpha function. We show that such networks can be successfully trained on noisy Boolean logic tasks and on the MNIST dataset encoded in time. We show that the spiking neural network outperforms comparable spiking models on MNIST and achieves similar quality to fully connected conventional networks with the same architecture. The spiking network spontaneously discovers two operating modes, mirroring the accuracy-speed trade-off observed in human decision-making: a highly accurate but slow regime, and a fast but slightly lower-accuracy regime. These results demonstrate the computational power of spiking networks with biological characteristics that encode information in the timing of individual neurons. By studying temporal coding in spiking networks, we aim to create building blocks towards energy-efficient, state-based and more complex biologically-inspired neural architectures.

Accurate modelling of the radiation output of a medical linear accelerator is important for radiotherapy treatment planning. The major challenge is the adjustment of the model to a specific treatment unit. One approach is to use a multiple-source model containing a set of physical parameters. In this work, the parameters were derived from standard beam data measurements using optimization methods. The source model used includes sub-sources for bremsstrahlung radiation from the target, extra-focal photon radiation and electron contamination. The cost function includes a gamma error measure between measurements and current dose calculations. The procedure was applied to six beam data sets (6 MV to 23 MV) measured with accelerators from three vendors, but the results focus primarily on Varian accelerators. The obtained average gamma error (1%, 1 mm) between dose calculations and measurements used in optimization was smaller than 0.7 for each studied treatment beam and field size, and a minimum of 83% of measurement points passed the gamma < 1 criterion. For experiments made at different SSDs and for asymmetric fields, the average gamma errors were smaller than 1.1. For irregularly shaped MLC apertures, the differences in point doses were smaller than 1.0%. This work demonstrates that the source model parameters can be automatically derived from simple measurements using optimization methods. The developed procedure is applicable to a wide range of accelerators, and has an acceptable accuracy and processing time.

The development of a unique neurosurgical navigator is described and a preliminary series of seven cases of intracerebral lesions approached with the assistance of this neuronavigation system under ultrasound control is presented. The clinical series included five low-grade astrocytomas, one chronic intracerebral hematoma, and one porencephalic cyst. Management procedures included biopsy in all cases, drainage of the hematoma, and endoscopy and fenestration for the cyst. The features of the neuronavigation system are interactive reconstructions of preoperative computerized tomography and magnetic resonance imaging data, corresponding intraoperative ultrasound images, versatility of the interchangeable end-effector instruments, graphic presentation of instruments on the reconstructed images, and voice control of the system. The principle of a common axis in the reconstructed images served to align the navigational pointer, biopsy guide, endoscope guide, ultrasound transducer, and surgical microscope to the brain anatomy. Intraoperative ultrasound imaging helped to verify the accuracy of the neuronavigator and check the results of the procedures. The arm of the neuronavigation system served as a holder for instruments, such as the biopsy guide, endoscope guide, and ultrasound transducer, in addition to functioning as a navigational pointer. Also, the surgical microscope was aligned with the neuronavigator for inspection and biopsy of the hematoma capsule to rule out tumor etiology. Voice control freed the neurosurgeon from manual exercises during start-up and calibration of the system.