[Safety Analysis Using Event Tree Analysis for Multi-Ion Therapy].

At the National Institutes for Quantum Science and Technology (QST), a multi-ion therapy using helium, carbon, oxygen, and neon ions has been studied for charged particle therapy with more optimal biological effects. To make multi-ion therapy clinically feasible, a new treatment system was developed to realize the changes of the ion species in each irradiation using the Heavy Ion Medial Accelerator in Chiba (HIMAC). Since radiation therapy is safety-critical, it is necessary to construct a safety system that includes multiple safety barriers in the new treatment system for multi-ion therapy and to perform a safety analysis for the prevention of serious accidents. In this study, we conducted a safety analysis using event tree analysis (ETA) for newly introduced processes in the treatment planning, accelerator, and irradiation system of the multi-ion therapy. ETA is an optimal method to verify multiple safety barriers that are essential for medical safety and to shorten the time for safety analysis by focusing only on the new processes. Through ETA, we clarified the types of malfunctions and human errors that may lead to serious accidents in the new system for multi-ion therapy, and verified whether safety barriers such as interlock systems and human check procedures are sufficient to prevent such malfunctions and human errors. As a result, 6 initial events which may lead to serious accidents were listed in the treatment planning process, 16 initial events were listed in the accelerator system, and 13 initial events were listed in the irradiation system. Among these 35 initial events, 5 cautionary initial events were identified that could lead to serious final events and they had a probability of occurrence higher than 10-4. Meanwhile, the others were all initial events that do not lead to serious accidents, or the initial events that can lead to serious accidents but were considered to have sufficient safety barriers. The safety analysis using ETA successfully identified the system malfunctions and the human errors that can lead to serious accidents, and the multiple safety barriers against them were systematically analyzed. It became clear that the multiple safety barriers were not sufficient for some initial events. We plan to improve the safety barriers for the five cautionary initial events before the start of the clinical trial. Based on these findings, we achieved our objective to conduct a safety analysis for a new treatment system for multi-ion therapy. The safety analysis procedure using ETA proposed by this study will be effective when new systems for radiotherapy are established at QST and other facilities in the future as well.

Regional left atrial conduction velocity in the anterior wall is associated with clinical recurrence of atrial fibrillation after catheter ablation: efficacy in combination with the ipsilateral low voltage area.

BACKGROUND: Left atrial (LA) conduction velocity (CV) is an electrical remodeling parameter of atrial fibrillation (AF) substrate. However, the pathophysiological substrate of LA-CV and its impact on outcomes after catheter ablation for AF have not been well evaluated. METHODS: We retrospectively evaluated 119 patients with AF who underwent catheter ablation and electroanatomical mapping during sinus rhythm. To measure regional LA-CV, we took triplet sites (A, B, and C) on the activation map and calculated the magnitude of the matched orthogonal projection vector between vector-AB and vector-AC, indicating two-dimensional CV. The median of the LA-CVs from four triad sites in both the anterior and posterior walls was set as the 'model LA-CV'. We evaluated the impact of the model LA-CV on recurrence after ablation and relationship between the model LA-CV and LA-low voltage area (LVA) of < 0.5 mV. RESULTS: During the 12-month follow-up, 29 patients experienced recurrence. The LA-CV model was significantly correlated with ipsilateral LVA. The lower anterior model LA-CV was significantly associated with recurrence, with the cut-off value of 0.80 m/s having a sensitivity of 72% and specificity of 67%. Multivariable analysis revealed that the anterior model LA-CV (hazard ratio, 0.09; 95% confidence interval, 0.01-0.94; p = 0.043) and anterior LA-LVA (hazard ratio, 1.06; 95% confidence interval, 1.00-1.11; p = 0.033) were independently associated with AF recurrence. The anterior LA-LVA was mildly correlated with the anterior model LA-CV (r = -0.358; p < 0.001), and patients with both lower LA-CV and greater anterior LA-LVA based on each cut-off value had the worst prognosis. However, decreased LA-CV was more likely to be affected by the distribution pattern of the LVA rather than the total size of the LVA. CONCLUSION: Decreased anterior LA-CV was a significant predictor of AF recurrence and was a useful electrical parameter in addition to LA-LVA for estimating AF arrhythmogenicity.

Long P-wave duration immediately after pulmonary vein isolation on radiofrequency catheter ablation for atrial fibrillation predicts clinical recurrence: correlation with atrial remodeling in persistent atrial fibrillation.

P-wave morphology reflects atrial remodeling and indicates prognosis after radiofrequency catheter ablation (RFCA) for atrial fibrillation (AF). The impact of p-wave morphology after excluding the effect of pulmonary vein (PV) substrate on outcomes is unknown. We evaluated the p-wave morphology on electrocardiography immediately after PV isolation for clinical outcomes. Eighty-four consecutive patients (47 with paroxysmal AF and 37 with persistent AF) who underwent RFCA were included. P-wave duration (PWD) and amplitude in all leads were examined during sinus rhythm immediately after PV isolation. We evaluated the relationship between electrocardiogram parameters and AF recurrence, according to the type of AF and following ablation, and the correlation with left atrial (LA) volume, low voltage ratio, and fixed conduction time. During 12 months of follow-up, 20 patients experienced recurrence. The cut-off value of PWD > 120 ms in lead I showed a sensitivity of 75% and specificity of 69% for predicting recurrence. PWD was significantly correlated with LA volume, low voltage, and conduction velocity. Significantly higher recurrence rates were observed in patients with PWD > 120 ms than in those with PWD ≤ 120 ms (p < 0.001), and the difference was more pronounced in patients with persistent AF. Multivariate analysis demonstrated that PWD > 120 ms was independently associated with recurrence in the total population (hazard ratio 2.00; 95% confidence interval 1.27-3.22; p = 0.003) and in patients with persistent AF. In conclusion, long PWD after PV isolation predicts AF recurrence, which might be associated with the extent of the LA substrate in persistent AF.

A simple algorithm for beam profile diagnostics using a thermographic camera.

A new algorithm for digital image processing apparatuses is developed to evaluate profiles of high-intensity DC beams from temperature images of irradiated thin foils. Numerical analyses are performed to examine the reliability of the algorithm. To simulate the temperature images acquired by a thermographic camera, temperature distributions are numerically calculated for 20 MeV proton beams with different parameters. Noise in the temperature images which is added by the camera sensor is also simulated to account for its effect. Using the algorithm, beam profiles are evaluated from the simulated temperature images and compared with exact solutions. We find that niobium is an appropriate material for the thin foil used in the diagnostic system. We also confirm that the algorithm is adaptable over a wide beam current range of 0.11-214 µA, even when employing a general-purpose thermographic camera with rather high noise (ΔT(NETD) ≃ 0.3 K; NETD: noise equivalent temperature difference).

(11)CH4 molecule production using a NaBH4 target for (11)C ion acceleration.

Solid-state materials suitable for use as proton irradiation targets were investigated for producing high-purity (11)CH4 molecules for heavy-ion cancer therapy. The radioactivity of gas produced by proton irradiation was measured for several target materials. Also, the radioactive molecular species of the produced gas were analyzed by radio gas chromatography. We found that 5 × 10(12) (11)C molecules could be collected by proton irradiation on a NaBH4 target. We also found that the (11)CH4 molecules were produced and collected directly from the irradiated target, owing to the hydrogen atoms bound in the solid-state NaBH4.

Numerical studies of transient gain reduction process in a multi-wire proportional chamber.

A gain reduction process caused by successive beam irradiation in a multi-wire proportional chamber was numerically investigated to clarify the relations between the gas gain variation and the ion density distribution. A numerical code was developed based on a two-dimensional drift-diffusion model in order to evaluate the ion and electron density distributions and the electric field variation caused by the space charge effect. In order to consider the gain reduction process which occurs under the high rate and successive irradiation, the simulations were performed for the time period of ∼10-100 µs, which is much longer than the time required for ions to travel from an anode to a cathode. The numerical simulation results showed that for the low gas gain regime of ∼10, quasi-stationary density distribution of the ions was formed by the high-rate beams of ∼10(8)-10(10) particles per second, and that the transient variation of the gas gain became constant after establishment of the quasi-stationary ion density distributions.

Generation of a monoclonal antibody specific for tissue-nonspecific alkaline phosphatase and its use in a clinical diagnostic study.

Tissue-nonspecific alkaline phosphatase (TNSALP) in serum comprises liver alkaline phosphatase (liver-ALP) and bone alkaline phosphatase (bone-ALP). Liver-ALP is a marker of liver disease; thus a specific method for its measurement would be useful. Measurement of ALP by electrophoresis is difficult, although all of the isozymes can be assessed simultaneously. Total ALP can also be measured by automated analyzer, but it is difficult to determine the cause of a high ALP value because bone-, intestine-, placenta-, and tumor-ALP are measured together. Thus, anti-TNSALP monoclonal antibodies that can resolve these problems are needed. Here we have generated an anti-TNSALP monoclonal antibody, 3-29-3R. This clone has specificity to liver-ALP rather than to bone-ALP. In electrophoresis, 3-29-3R reacted with TNSALP and shifted the bands. The use of 3-29-3R enabled easy interpretation of the results. Furthermore, we tested 3-29-3R by developing an immunocapture enzymatic assay (IEA). Preliminary results of the IEA show that this method is effective for measurement of liver-ALP. Thus, the monoclonal antibody that we have established may be a useful tool for clinical diagnosis.

Development and characterization of novel monoclonal antibodies against tartrate-resistant acid phosphatase 5.

Serum band 5 tartrate-resistant acid phosphatase (TRACP 5; EC 3.1.3.2) is a glycoprotein that exists as two very similar isoforms, TRACP 5a and TRACP 5b. The similarity of these two isoforms has made it difficult to establish monoclonal antibodies specific for either isoform. We report here the development of a monoclonal antibody with high specificity for TRACP 5b. We prepared TRACP 5b antigens from four sources: TRACP 5b purified from human bone, recombinant TRACP 5 from Escherichia coli, recombinant TRACP 5 from insect cells, and a synthetic TRACP 5b peptide. Thirty-seven mice were each immunized with 1 of the 4 different TRACP antigens to generate 473 antibody-producing clones. Three of these clones, Trk27, Trk49, and Trk62, reacted with TRACP 5b. These three clones were all established from mice exposed to native bone TRACP 5b antigen. In fact, none of the other antigens were able to generate anti-TRACP 5b monoclonal antibodies in mice. Furthermore, Trk62 interacted more strongly with TRACP 5b than with TRACP 5a. These results suggested that although recombinant proteins can be effective antigens, the native TRACP 5 protein might be more effective at generating monoclonal antibodies of greater specificity due to its more faithful representation of the native three-dimensional structure of the protein.

Multiple model-based reinforcement learning.

We propose a modular reinforcement learning architecture for nonlinear, nonstationary control tasks, which we call multiple model-based reinforcement learning (MMRL). The basic idea is to decompose a complex task into multiple domains in space and time based on the predictability of the environmental dynamics. The system is composed of multiple modules, each of which consists of a state prediction model and a reinforcement learning controller. The "responsibility signal," which is given by the softmax function of the prediction errors, is used to weight the outputs of multiple modules, as well as to gate the learning of the prediction models and the reinforcement learning controllers. We formulate MMRL for both discrete-time, finite-state case and continuous-time, continuous-state case. The performance of MMRL was demonstrated for discrete case in a nonstationary hunting task in a grid world and for continuous case in a nonlinear, nonstationary control task of swinging up a pendulum with variable physical parameters.