Biomedical Engineering: Applications, Basis and Communications

Invasive Ductal Carcinoma (IDC) is one of the most frequently diagnosed breast cancers. IDC accounts for about 8 out of 10 of all invasive breast cancers. While early detection of breast cancer is essential for the reduction of death rate, there may be already more than 10⁷ cells in a breast cancer when it can be observed by X-ray mammogram. In contrast, the passive IR spectrogram proposed by Szu et al. was shown to be promising in detecting breast cancers several months ahead of mammogram. With energy readings from two IR cameras, middle wavelength IR (MIR, 3–5 μm) and long wavelength IR (LIR, 8–12 μm), dual-spectrum IR (DS-IR) spectrogram may be computed by using the deterministic neighborhood-based blind source separation algorithm developed by Szu et al..^4–7 To evaluate the performance of the DS-IR spectrogram on detection of IDC, a DS-IR spectrogram hardware system is built and a sub-pixel super-resolution registration is developed to implement the deterministic neighborhood-based blind source separation algorithm. Clinical tests have been carried out with the approval of Institutional Review Board of National Taiwan University Hospital. From August 2007 to June 2008, 35 patients aged between 30–66 (average age 49) with IDC breast cancers were recruited in this project. The results demonstrate that 62.86% of success rate for IDC detection may be achieved with the cross-sectional data. Longitudinal study shows that breast cancers may be detected more accurately by cross-referencing s₁ maps of multiple time-points.

Compartment syndrome (CS) is the pressure increasing within a confined anatomic space, which results in ischemia to the encompassed tissues and then leads to cells death. If left untreated, CS results in irreversible cell damage, further permanently postischemic scarring and contracture. The intracompartmental pressure is a major diagnostic consideration. Various methods of compartment pressure (CP) measurement have been developed since the late 1800s. They can be classified into invasive and noninvasive categories. Invasive methods are accurate but painful; whereas noninvasive approaches are more comfortable but less accurate. Both methods provide reproducible outcomes but also require specific CP measurement instrument to be performed. This research proposes a rapid and convenient method via examining ultrasonographic fascia wall displacement. For the viscoelastic nature of the compartment fascia, our experiment results have shown the ultrasonographic fascia displacement accurately reflects even small changes of the compartment diameter. There is a strong correlation between compartment pressure and fascia displacement ratio. It also recommends that the ratio of 1.4 could be chosen as the critical value for determining acute compartment syndrome. These results suggest that ultrasonography may serve as an adequate noninvasive tool to monitor CP variations.

Image based registration of rigid objects has been frequently addressed in the literature to obtain an object's motion parameters. In this paper, a new approach of joint segmentation-rigid registration, within the variational framework of the phase field approximation of the Mumford-Shah's functional, is proposed. The defined functional consists of two Mumford-Shah equations, extracting the discontinuity set of the reference and target images due to a rigid spatial transformation. Multiscale minimization of the proposed functional after finite element discretization provided a sub-pixel, robust algorithm for edge extraction as well as edge based rigid registration. The implementation considerations of the proposed method, including memory usage, convergence rate and effects of parameters selection, was discussed and its efficacy was examined in a comprehensive set of synthetic, phantom and clinical experiments. It was found that the proposed joint segmentation-rigid registration method provides improved results, in comparison with the currently available methods which are often based on maximizing images similarities, especially when the reference and target images are of different qualities. A high registration accuracy was obtained when estimating the knee joint kinematics through MR images taken at different joint configurations. It was concluded that the proposed method can be effectively used in applications where 3D image registration of rigid objects is concerned, e.g. for estimation of the motion parameters of human joints.

This paper explores grasping in robot-aided upper extremity rehabilitation, with a special focus on reaching and grasping exercises and the coordination between load force and grasp force. Six healthy subjects and two hemiparetic subjects performed "pick and place" movements with a haptic robot and virtual environment. These movements were segmented into three phases: grasping, transport and release phase, and the correlation between grasp and load force was calculated over the entire movement and within each phase separately. Results show that the subjects employ same basic mechanism of grasp and load force coordination during a virtual task as in real situations. However, the grasp and load force are partially decoupled due to the nature of the grasping device and the complexity of the task. Furthermore, the coordination is different in different phases and also depends on the level of impairment as well as the level of active support by the rehabilitation robot. The first hemiparetic subject, who can perform reaching movements but cannot open the hand, thus has a lower correlation between grasp force and load force than healthy subjects only in the release phase while the second hemiparetic subject, who has little arm mobility, has a lower correlation in all three phases. Thus, the current work provides basic empirical knowledge that can serve as a basis for future research and for the design of robot-aided reaching and grasping tasks.

The malignant ventricular tachyarrhythmia including ventricular tachycardia (VT) and ventricular fibrillation (VF) is the major cause of triggering sudden cardiac death (SCD) and it is seriously harmful to human. There is a great significance to predict the VT/VF. In this study, the RR interval series preceding the onset of VT/VF events are used as the study objects, and the 135 RR interval series are recorded by implantable cardioverter defibrillators (ICD) from 78 patients. Instantaneous heart rate (IHR) series are obtained after preprocessing the RR interval series. The Hilbert spectrum and the frequency marginal spectrum of IHR series are analyzed based on the traditional Hilbert-Huang transform (HHT) and the improved HHT. Some signal spectrum features are extracted from the marginal spectrum of IHR series: low frequency amplitude, high frequency amplitude, very high frequency amplitude, total amplitude and the low-to-high frequency amplitude ratio. The statistical analysis shows that the performance of improved HHT is better than that of traditional HHT. High frequency amplitude, very high frequency amplitude and total amplitude of IHR series preceding the onset of VT/VF are significantly higher than that of normal sinus rhythm (p < 0.003), and low-to-high frequency amplitude ratio is significantly lower than that of normal sinus rhythm (p < 0.0002).

For the past few years, the medical image registration technology has made rapid and significant progress, and has been extensively applied for the 2D/3D medical image processing. However, the robustness of the similarity metric in the medical image registration is rarely studied. In this paper, the mutual information-based registration technology is introduced and the concept of the robustness of the similarity metric is defined. The robustness of the mutual information similarity metric is analyzed and discussed from three aspects such as interpolation methods, image data loss and noise corruption after the linear, quadratic spline, cubic spline, quadratic B-spline, and cubic B-spline interpolations are elaborated and studied. The robustness experiments reveal that the mutual information similarity metric can obtain good robustness in the case of the use of various interpolation methods in the medical image registration; the mutual information similarity metric can also keep good robustness in the case of a slight loss of image data. However, the metric will fail to register the images on the condition that the medical images are seriously incomplete. In addition, we find, if the medical images corrupted by the salt & pepper noise, then the metric basically can succeed in aligning the medical images; but regretfully, the metric fully fail under the condition of the medical images corrupted by the Gaussian noise.

Brackets are basic elements of orthodontic systems. Traditionally, they are produced by a powder-injection mass production process, by which it is difficult to develop individualized products. This study applies computer-aided engineering techniques, including digitizing, virtual model design, and additive manufacturing (AM) to orthodontics for the production of customized brackets. First, a digital model of the teeth was obtained via a 3D scanning process, in which an individual teeth plaster mold was digitized; customized brackets were then designed based on the model, and physical brackets were rapidly fabricated by AM processing. Finally, key dimensional measurements of the brackets were taken to confirm whether the brackets fulfilled the functional requirements for clinical usage. By this process, brackets designed for different teeth pads were obtained by digital planning, CAD design and rapid manufacturing. The custom-made brackets include complete dental profile information, and increase the efficiency of treatment, save clinical setup time, and reduce the treatment duration and cost for the patient. The direct manufacture of metal brackets by an AM process is a modern method with great potential to be used in the generation of customized brackets for orthodontic treatment.

Planar electrode array is an important tool to evaluate perceptual or cognitive functions of the cortex and prosthetic applications. Many construction methods have been developed. To maximize the usefulness of an array electrode, a low-cost, precise, and flexible microelectrode array with low man power and short construction duration is crucial. In this study, we introduced an 8 × 8 microelectrode array on a flexible polyimide film through microelectronics fabrication. The array dimension was capable of covering the primary somatosensory cortex of a rat. The microelectrode array was insulated with biocompatible Parylene-C except of microelectrode tip. Each electrode tip was 66 μm height and separated with 0.5 mm to refine a detail somatic sensory processing. In pentobarbital anesthetized rats, stable spontaneous brain activity was successfully recorded through the electrode array. In addition, positive peaks of somatosensory evoked potentials (SEPs) elicited by stimulating rat's whisker pad, forepaw, hindpaw, and tail were obviously and consistently recorded. Latencies of SEPs increased as caudal part of the body was stimulated. The SEPs from stimulation of 4 body parts revealed different spatiotemporal patterns, which indicated a somatotopic organization of the rat. Our results demonstrated the superiority of the planar microelectrode array on the application of simultaneous recording and analysis of the brain activity in rats.

Children with cerebral palsy (CP) have been reported to have various levels of deficits in balance control, which can be described using the relationship between the body's centre of mass (COM) and the centre of pressure (COP). This study aimed to investigate the balance control of children with spastic diplegic CP during level walking. The COM-COP inclination angles and angular velocities, as well as temporal-spatial variables from 12 children with spastic diplegic CP (seven girls and five boys, aged 12.4 ± 4.4 years) and 12 normal controls (eight girls and four boys, aged 11.2 ± 4.4 years) were obtained using a motion analysis system and two forceplates. With compromised balance control as a result of neuromusculoskeletal pathologies, the CP group walked with reduced walking speed and stride length (p < 0.05), but increased stride time and step width (p < 0.05), indicating reduced gait efficiency. They also showed significantly reduced anterioposterior COM-COP inclination angles and angular velocities (p < 0.05), but increased mediolateral COM-COP inclination angles and angular velocities (p < 0.05) when compared to the normal controls. The latter phenomenon may be related to an increased risk of falling in these patients. Therefore, it appears that programs and/or devices for preventing falls are needed for children with spastic diplegic CP.

Survival from cardiac arrest is dependent on timely cardiopulmonary resuscitation (CPR). Experimental and modeling work had shown that the relationship between compression force and sternal displacement had a tendency of hysteresis curve, which in manikins was rather lineal. A novel method was introduced to improve the mechanical characteristics of the manikins using a variable-stiffness springs group-damper structure, in which the spring's group and the damper simulate the elastic and damping of human chest respectively. To do the simulation, the model of the human chest's mechanical during CPR based on Gruben's pre-result was modified and the elastic part was fitted by piecewise linear function to get the springs' stiffness. The variable-stiffness springs group system was designed accordingly to simulate the human chest' stiffness during CPR, and a damper was designed to simulate the damping of chest. The damper and the variable-stiffness springs group were paralleling combined, forming a mechanical system. A sample system was realized and the test results showed that there were nonlinear elasticity and adequate viscosity in this system, whose coefficients were adjacent to Gruben's 'typical' human. According to the method, the mechanical system could be changed to accurately simulate different models, and its stiffness could be easily adjusted by varying the interval between the nest springs' top. The simulation method could be used directly in manikin, whose mechanical character would be improved and then more adequate training would be provided.

This study compared cervical and ocular vestibular-evoked myogenic potentials (cVEMPs and oVEMPs) between air-conducted sound (ACS) and bone-conducted vibration (BCV) modes to determine whether these two stimulation modes activate the same population of primary vestibular afferents. Fifteen healthy subjects underwent cVEMP and oVEMP tests using ACS stimuli at 127 dB pe SPL and BCV stimuli at 128 dB force level. The characteristic parameters of cVEMPs and oVEMPs were compared between ACS and BCV modes. The mean p13 and n23 latencies of ACS-cVEMPs were significantly longer than those of BCV-cVEMPs. Likewise, the mean nI and pI latencies for ACS-oVEMPs were also significantly longer than those for BCV-oVEMPs. There was no significant difference in the mean amplitude of cVEMPs between the ACS and BCV modes. However, comparing the oVEMP amplitude, a relationship: (Amplitude of BCV-oVEMP) = 2.3 x (Amplitude of ACS-oVEMP) was demonstrated. In conclusion, the population of primary vestibular afferents activated by ACS and BCV stimuli is similar for cVEMPs. In contrast with oVEMPs, BCV mode activates more number of primary vestibular afferents than ACS mode does. In interpreting oVEMP and cVEMP results, stimulation mode should be checked first.