Y. Louhisalmi

Development of a robotic surgical assistant

After development of a passive articulated localization arm we are now developing a robotic surgical assistant for neurosurgery. Our study includes ergonomic, precision and safety aspects. As a result of early work interactive hand and computer driven use might be developed. Mechatronic approach is supposed to minimize systems weight, friction and inertia of mass for user. The goal of this work, interests and preliminary results are presented.

Neuronavigator-Guided Cerebral Biopsy

Neuronavigators are new dynamic interactive instruments that use on-line computers to orient imaging data to the surgical field and guide the neurosurgeon to his target. We have been working since 1987 on a neuronavigator that serves not only as a precise pointer, but also as a dynamic arm that can be used to hold instruments, such as biopsy guides. The neuronavigator arm consists of six joints with optical encoders and is attached to the Mayfield headholder. The arm is connected to a workstation running customized 3D image graphics software. Special instruments and surgical technique have been developed. Here, we report on early clinical experience with ten biopsy procedures: 4 low-grade and 3 high-grade astrocytomas, one craniopharyngioma and one chronic intracerebral haematoma and intracerebral cyst, both of the latter with surrounding tumour suspect tissue. In all glioma cases serial biopsies were taken from optimal sites under ultrasound imaging control. Eight cases showed representative tumour tissue, while in two cases neoplasia was ruled out. The neuronavigator proved to be versatile, allowing comprehensive imaging data to be adapted to the surgical field.

Computer-guided Laser For Neurosurgery

On the basis of over 40 neurosurgical laser operations, including CO2, Nd-YAG and simultaneous CO2/Nd-YAG laser procedures, a computer-guided system for spatial control of the laser beam has been developed. The pilot laser has several modes: it can direct the neurosurgeon along the central axis of the surgical microscope to stereotactically determined point-like targets or outline selected layers of underlying volume targets onto superficial surfaces such as scalp and cortex and onto the tissue at the appropriate depth. The active treatment laser can be guided by preoperative CT/MRI or intraoperative ultrasound image data for layer-by-layer resection of tumor. The laser system can be connected to the surgical field by rigid stereotactic means or by neuronavigator. In the present system, a special brain surgery adapter coordinates the imaging system and laser to the surgical field. Thus, the laser system can be used for image-guided surgical orientation, for demarcation of lesions and for actual layer-by-layer removal of tumor.

Development of an accuracy assessment phantom for surgical navigators

The objective of this study was to design a calibration phantom for a surgical navigator used in a hospital environment. It addresses two major issues: the design of an accuracy phantom and the accuracy analysis of the surgical navigator in a hospital setting. The designed phantom was used to assess the accuracy of the optical tracking modality of the surgical navigator used at Oulu University Hospital, Oulu, Finland. The phantom functioned according to the design criteria, it was easy to use and it had enough calibration points that were localized by the navigator according to the accuracy assessment protocol to assess the accuracy error. The distances measured from a fixed origin with the surgical navigator were compared to the known phantom calibration point coordinates. The mean error was within the manufacturer specifications of 1.00 mm. The analysis done using the designed phantom and accuracy assessment protocol showed that the error increased with the distance from the center of the phantom. The accuracy assessment protocol using the present phantom proved to be a suitable method for accuracy analysis of a surgical navigator in a hospital setting.

Neurosurgical Navigation System

A mechatronic system for neurosurgical simulation before operation as well as for orientation during surgery is presented. The system consists of a six-jointed robotic arm, workstation and customized 3D visualization software. The robotic arm is used to rigidly transfer multiple image data to the surgical volume. The system has been used in several biopsy and other intracranial operations with good results.

Design of visualization system for neurosurgical workstation

A visualization system for simulating nellrosurgical operations and helping in the orientation during neurosurgical procedures is presented. The system generates 3D and 2D reconstructions from preopertive M R and CT images. These reconstrllctions help in conjunction with intraoperative ultrasound the neurosurgeon in localizing the target in the brain tissue before craniotomy and during the opertion. The system is controlled by a mouse or speech in preoperative planning, while during the operation the neurosurgeon needs to use only voice control tlws leaving hands free. The flexible, expandable and portable design allows the visualization system to grow with the neurosurgical workstation.

Development of a localization arm for neurosurgery

Basic requrements of a useful localization system for neurosurgery have been formed and presented. A localization arm has been realized fulfilling the requirements. The first version of the arm was undergone laboratory tests and the second version is now under clinical testing.

Image transformation from polar to Cartesian coordinates simplifies the segmentation of brain images

A new image transformation has been designed to simplify the segmentation of the MR and CT brain images. This transformation supposes that the head is a cylindrical object and can be modeled more efficiently in polar coordinates, which gives a new set of geometrically constant image features. For example, the air-skin surface can be modeled with a one-dimensional function. An imagt transformation from the polar coordinate system to the Cartesian coordinate system has been used, because the algorithm rea ization and the actual image processing are easier and faster in Cartesian coordinates and the results equal to polar coordinate processing.

Intraoperative feedback for neurosurgical stereotactic laser system

A method for accurate intraoperative guidance of a surgical laser beam in neurosurgery is discussed. This method uses intraoperative ultrasound imaging and computer processing to control the movement of the laser. The method is described and images are shown. The system has been tested using data from intraoperative ultrasound imaging of cerebral tumors and simulated tumors of various shapes.<<ETX>>

Design of High Robust Voice Interface for Speech Activated Neurosurgical Workstation

Several methods were developed in order to improve reliability and robustness of a voice interface designed for a speech activated neurosurgical workstation that is a command-number-limited system but directly involved in nurosurgical operations. We chose a commercial voice recognizer and synthesizer as basic voice environment of the voice interface. A parallel connection grammar structure in the command recognition and a practical operation procedure oriented logical structure in the command understanding were designed to remove misrecognition and misunderstanding. In the voice recognizing process, by means of setting noise pits and dynamically adjusting tolerance level, the substitution errors caused by uncorrected voice commands and noises in operating room could be greatly reduced. To guarantee the high robust voice control, we employed real-time voice feedback to confirm the accepted commands. Our testing results showed that employing these methods greatly improved voice interface robustness.