RADBIOMOD: A simple program for utilising biological modelling in radiotherapy plan evaluation

PurposeRadiotherapy plan evaluation is currently performed by assessing physical parameters, which has many limitations. Biological modelling can potentially allow plan evaluation that is more reflective of clinical outcomes, however further research is required into this field before it can be used clinically.MethodsA simple program, RADBIOMOD, has been developed using Visual Basic for Applications (VBA) for Microsoft Excel that incorporates multiple different biological models for radiotherapy plan evaluation, including modified Poisson tumour control probability (TCP), modified Zaider–Minerbo TCP, Lyman–Kutcher–Burman normal tissue complication probability (NTCP), equivalent uniform dose (EUD), EUD-based TCP, EUD-based NTCP, and uncomplicated tumour control probability (UTCP). RADBIOMOD was compared to existing biological modelling calculators for 15 sample cases.ResultsComparing RADBIOMOD to the existing biological modelling calculators, all models tested had mean absolute errors and root mean square errors less than 1%.ConclusionsRADBIOMOD produces results that are non-significantly different from existing biological modelling calculators for the models tested. It is hoped that this freely available, user-friendly program will aid future research into biological modelling.     

Two-dimensional imaging of tumour control probabilities and normal tissue complication probabilities

Aim

To create a presentation method of TCP and NTCP distributions calculated based on dose distribution for a selected CT slice.

Materials and methods

Three 24-bit colour maps – of dose distribution, delineated structures and CT information – were converted into m-by-n-by-3 data arrays, containing intensities of red, green, and blue colour components for each pixel. All calculations were performed with Matlab v.6.5. The transformation function, which consists of five linear functions, was prepared to translate the colour map into a one-dimensional data array of dose values. A menu-driven application based on the transformation function and mathematical models of complication risk (NTCP) and treatment control probability (TCP) was designed to allow pixel-by-pixel translation of colour maps into one-dimensional arrays of TCP and NTCP values.

Results

The result of this work is an application created to visualize the TCP and NTCP distribution for a single CT scan based on the spatial dose distribution calculated in the treatment planning system. The application allows 10 targets (PTV) and 10 organs at risks (OaR) to be defined. The interface allows alpha/beta values to be inserted for each delineated structure. The application computes TCP and NTCP matrices, which are presented as colour maps superimposed on the corresponding CT slice. There is a set of parameters used for TCP/NTCP calculations which can be defined by the user.

Conclusion

Our application is a prototype of an evaluation tool. Although limited to a single plane of the treatment plan, it is believed to be a starting point for further development.     

A radiobiological comparison of hypo-fractionation versus conventional fractionation for breast cancer 3D-conformal radiation therapy

BackgroundThe present research was aimed to compare the toxicity and effectiveness of conventional fractionated radiotherapy versus hypo-fractionated radiotherapy in breast cancer utilizing a radiobiological model.Materials and methodsThirty-five left-sided breast cancer patients without involvement of the supraclavicular and axillary lymph nodes (with the nodal stage of N0) that had been treated with conventional or hypo-fractionated were incorporated in this study. A radiobiological model was performed to foretell normal tissue complication probability (NTCP) and tumor control probability (TCP).ResultsThe data represented that TCP values for conventional and hypo-fractionated regimens were 99.16 ± 0.09 and 95.96 ± 0.48, respectively (p = 0.00). Moreover, the NTCP values of the lung for conventional and hypo-fractionated treatment were 0.024 versus 0.13 (p = 0.035), respectively. Also, NTCP values of the heart were equal to zero for both regimens.ConclusionIn summary, hypo-fractionated regimens had comparable efficacy to conventional fraction radiation therapy in the case of dosimetry parameters for patients who had left breast cancer. But, utilizing the radiobiological model, conventional fractionated regimens presented better results compared to hypo-fractionated regimens.     

Normal tissue complication probability models for severe acute radiological lung injury after radiotherapy for lung cancer

PurposeTo derive Normal Tissue Complication Probability (NTCP) models for severe patterns of early radiological radiation-induced lung injury (RRLI) in patients treated with radiotherapy (RT) for lung tumors. Second, derive threshold doses and optimal doses for prediction of RRLI to be used in differential diagnosis of tumor recurrence from RRLI during follow-up.Methods and materialsLyman-EUD (LEUD), Logit-EUD (LogEUD), relative seriality (RS) and critical volume (CV) NTCP models, with DVH corrected for fraction size, were used to model the presence of severe early RRLI in follow-up CTs. The models parameters, including α/β, were determined by fitting data from forty-five patients treated with IMRT for lung cancer. Models were assessed using Akaike information criterion (AIC) and area under receiver operating characteristic curve (AUC). Threshold doses for risk of RRLI and doses corresponding to the optimal point of the receiver operating characteristic (ROC) curve were determined.ResultsThe α/βs obtained with different models were 2.7–3.2 Gy. The thresholds and optimal doses curves were EUDs of 3.2–7.8 Gy and 15.2–18.1 Gy with LEUD, LogEUD and RS models, and μ_d of 0.013 and 0.071 with the CV model. NTCP models had AUCs significantly higher than 0.5. Occurrence and severity of RRLI were correlated with patients’ values of EUD and μ_d.ConclusionsThe models and dose levels derived can be used in differential diagnosis of tumor recurrence from RRLI in patients treated with RT. Cross validation is needed to prove prediction performance of the model outside the dataset from which it was derived.     

Dosimetric and radiobiological comparison in head-and-neck radiotherapy using JO-IMRT and 3D-CRT

IntroductionDosimetric and radiobiological evaluations for the Jaws-only Intensity-modulated radiotherapy (JO-IMRT) technique for head and neck jaws-only intensity-modulated radiation therapy (JO-IMRT) and 3D conformal radiation therapy (3D-CRT). To compare the head-and-neck therapeutic approaches utilizing JO-IMRT and 3D-CRT techniques, different radiation dose indices were calculated, including: conformity index (CI), homogeneity index (HI), and radiobiological variables like Niemierko's equivalent uniform dose based tumor control probability (TCP) of planning target volume (PTV), normal tissue complication probability (NTCP) of organs at risk (OAR) (brainstem, spinal cord, and parotid grand).Materials and methodsTwenty-five nasopharynx patients were studied using the Prowess Panther Treatment Planning System (Prowess Inc). The results were compared with the dose distribution obtained using 3D-CRT.ResultsRegarding tumor coverage and CI, JO-IMRT showed better results than 3D-CRT. The average doses received by the PTVs were quite similar: 72.1 ± 0.8 Gy by 3D-CRT and 72.5 ± 0.6 Gy by JO-IMRT plans (p > 0.05). The mean doses received by the parotid gland were 56.7 ± 0.7 Gy by 3D-CRT and 26.8 ± 0.3 Gy by JO-IMRT (p > 0.05). The HI and CI were 0.13 ± 0.01 and 0.14 ± 0.05 and (p > 0.05) by 3D-CRT and 0.83 ± 0.05 and 0.73 ± 0.10 by JO-IMRT (p < 0.05). The average TCP of PTV was 0.82 ± 0.08 by 3D-CRT and 0.92 ± 0.02 by JO-IMRT. Moreover, the NTCP of the parotid glands, brain stem, and spinal cord were lower using the JO-IMRT than 3D-CRT plans. In comparison to the 3D-CRT approach, the JO-IMRT technique was able to boost dose coverage to the PTV, improve the target's CI and HI, and spare the parotid glands. This suggests the power of the JO-IMRT over 3D-CRT in head-and-neck radiotherapy.     

Complication probability model for subcutaneous fibrosis based on published data of partial and whole breast irradiation

PurposeTo extend the application of current radiation therapy (RT) based normal tissue complication probability (NTCP) models of radiation-induced fibrosis (RIF) of the breast to include the effects of fractionation, inhomogeneous dose, incomplete recovery, and time after the end of radiotherapy in partial breast irradiation (PBI).Materials and methodsAn NTCP Lyman model with biologically effective uniform dose (BEUD) with and without a correction for the effect of incomplete repair was used. The time to occurrence of RIF was also taken into account. The radiobiological parameters were determined by fitting incidences of moderate/severe RIF in published randomized studies on RT of the breast. The NTCP model was used to calculate the risk of toxicity in 35 patients treated with intensity modulated, non-accelerated PBI and the result was compared with observed incidence of RIF.ResultsWith α/β fixed at 3Gy the parameters of the model without correction for incomplete repair extracted from fitting were: 50% complication probability biologically effective dose BEUD₅₀ = 107.2 Gy (95%CI = 95.9–118.6 Gy), volume parameter n = 0.06 (95%CI = 0–0.23), and slope of dose response m = 0.22, (95%CI = 0.20–0.23). After including the correction for incomplete repair with repair halftime for subcutaneous tissue of τ = 4.4 h we obtained BEUD₅₀ = 105.8 Gy (95%CI = 96.9–114.6Gy), n = 0.15 (95%CI = 0–0.33), m = 0.22 (95%CI = 0.20–0.23). Average NTCP predicted by these models, 4.3% and 2.0% respectively, offered a good agreement with RIF incidence in our patients, 5.7%, after an average follow-up of 12 months.ConclusionThe NTCP models of RIF, incorporating the effects of fractionation, volume effect, and latency of toxicity look promising to model PBI. Clinical validation from a prospective PBI treatment study is under development and will help test this preliminary result.     

The Use of Biologically Related Model (Eclipse) for the Intensity-Modulated Radiation Therapy Planning of Nasopharyngeal Carcinomas

Purpose

Intensity-modulated radiation therapy (IMRT) is the most common treatment technique for nasopharyngeal carcinoma (NPC). Physical quantities such as dose/dose-volume parameters are used conventionally for IMRT optimization. The use of biological related models has been proposed and can be a new trend. This work was to assess the performance of the biologically based IMRT optimization model installed in a popular commercial treatment planning system (Eclipse) as compared to its dose/dose volume optimization model when employed in the clinical environment for NPC cases.

Methods

Ten patients of early stage NPC and ten of advanced stage NPC were selected for this study. IMRT plans optimized using biological related approach (BBTP) were compared to their corresponding plans optimized using the dose/dose volume based approach (DVTP). Plan evaluation was performed using both biological indices and physical dose indices such as tumor control probability (TCP), normal tissue complication probability (NTCP), target coverage, conformity, dose homogeneity and doses to organs at risk. The comparison results of the more complex advanced stage cases were reported separately from those of the simpler early stage cases.

Results

The target coverage and conformity were comparable between the two approaches, with BBTP plans producing more hot spots. For the primary targets, BBTP plans produced comparable TCP for the early stage cases and higher TCP for the advanced stage cases. BBTP plans reduced the volume of parotid glands receiving doses of above 40 Gy compared to DVTP plans. The NTCP of parotid glands produced by BBTP were 8.0±5.8 and 7.9±8.7 for early and advanced stage cases, respectively, while those of DVTP were 21.3±8.3 and 24.4±12.8, respectively. There were no significant differences in the NTCP values between the two approaches for the serial organs.

Conclusions

Our results showed that the BBTP approach could be a potential alternative approach to the DVTP approach for NPC.     

The impact of PET/CT scanning on the size of target volumes,radiation exposure of organs at risk,TCP and NTCP,in the radiotherapy planning of non-small cell lung cancer

Aim

To compare radiotherapy plans made according to CT and PET/CT and to investigate the impact of changes in target volumes on tumour control probability (TCP), normal tissue complication probability (NTCP) and the impact of PET/CT on the staging and treatment strategy.

Background

Contemporary studies have proven that PET/CT attains higher sensitivity and specificity in the diagnosis of lung cancer and also leads to higher accuracy than CT alone in the process of target volume delineation in NSCLC.

Materials and methods

Between October 2009 and March 2012, 31 patients with locally advanced NSCLC, who had been referred to radical radiotherapy were involved in our study. They all underwent planning PET/CT examination. Then we carried out two separate delineations of target volumes and two radiotherapy plans and we compared the following parameters of those plans: staging, treatment purpose, the size of GTV and PTV and the exposure of organs at risk (OAR). TCP and NTCP were also compared.

Results

PET/CT information led to a significant decrease in the sizes of target volumes, which had the impact on the radiation exposure of OARs. The reduction of target volume sizes was not reflected in the significant increase of the TCP value. We found that there is a very strong direct linear relationship between all evaluated dosimetric parameters and NTCP values of all evaluated OARs.

Conclusions

Our study found that the use of planning PET/CT in the radiotherapy planning of NSCLC has a crucial impact on the precise determination of target volumes, more precise staging of the disease and thus also on possible changes of treatment strategy.     

A Simple Intervention for Improving the Implementation Rate of a Recommended Osteoporosis Treatment After Hip Fracture

ObjectiveOsteoporosis is often under-treated, and hip fracture is frequently its first manifestation. Hospitalization for a hip fracture is an opportunity to initiate osteoporosis treatment. The aim of this study was to investigate whether a simple intervention improves the implementation rate of a recommended osteoporosis treatment.MethodsOne hundred elderly patients admitted with low-impact hip fracture were given a 10 minute explanation about osteoporosis and its treatment during their postoperative hospital stay. In addition, the patients received an explanatory brochure and a letter to their primary care physician that included an article on fracture rate reduction with osteoporosis treatment. Implementation of therapy was assessed by a telephone survey 3 and 6 months postoperatively. The patients who had not received treatment at 3 months were given a repeated explanation. The historical control group was comprised of 100 hip fracture patients with similar demographic characteristics, who were operated on and discharged with the standard care recommendations for osteoporosis prevention.ResultsAt the 3 month follow-up, the therapy rate in both groups was similar (19%). Fifty-eight percent of the patients in the study group had no recollection of the intervention. However, after a repeated explanation, at the 6 month follow-up, 39% of the intervention group had received drug therapy for fracture prevention (P<.001).ConclusionA simple intervention enlisting the patients' help to involve their primary care physician can increase treatment rates for osteoporosis following a hip fracture. During the immediate postoperative period, the patients and their families have difficulty implementing the recommendations. Therefore, repeated communications are recommended. (Endocr Pract. 2013;19:46-50)     

A method to visualize the uncertainty of the prediction of radiobiological models

A method for quantitative visualization of the uncertainty in the predicted tumor control probability (TCP) and normal tissue complication probability (NTCP) in radiotherapy has been developed. Uncertainties of TCP and NTCP due to inter-individual variation of the underlying radiosensitivity parameters was simulated by sampling the prescribed dose from a uniform distribution and the radiosensitivity-parameters from a Gaussian distribution. The result is visualized as a scatter-plot superimposed to the population-based dose response curves using the prescribed dose as the common dosimetric variable. In addition, probability histograms are derived quantifying the probability of specific TCP- or NTCP-values for individual patients from the underlying population. The method is exemplified with a pleural mesothelioma case with the lung as organ at risk. A prescribed dose of 54 Gy together with radiosensitivity variations of 6% (tumor) and 10% (normal tissue) results in a TCP of 85% (range 68–94%, 90% confidence interval, CI) and an NTCP of 4% (range 3–6%, 90% CI), respectively. Increasing the radiosensitivity variation of the tumor to 15% and reducing the lung tolerance dose by 25% results in values of 84% (range 51–97%, 90% CI) for TCP and 9% (range 6–12%, 90% CI) for NTCP. Increasing the dose to 60 Gy leads to TCP- and NTCP-values of 93% (range 69–100%, 90% CI) and 12% (range 8–17%, 90% CI), respectively. The new method visualizes the uncertainty of TCP- and NTCP-values and hence of the therapeutic window. This can help the clinician to assess the treatment plan for the individual patient.     

盐酸美金刚联合银杏叶提取物对阿尔茨海默病患者血清BDNF、NGF、DA水平和认知功能的影响

目的:探讨盐酸美金刚联合银杏叶提取物对阿尔茨海默病患者血清脑源性神经营养因子(BDNF)、神经营养因子(NGF)浓度、多巴胺(DA)水平和认知功能的影响。方法:将我院2017年6月至2018年7月收治的92例阿尔茨海默病患者按随机数字表法分为对照组49例和研究组43例,对照组采用盐酸美金刚治疗,研究组在对照组基础上联合银杏叶提取物治疗。比较两组临床疗效,治疗前后血清BDNF、NGF、DA水平、认知功能、简易精神状态检查量表(MMSE)、日常生活能力(ADL)评分的变化和不良反应的发生情况。结果:治疗后,研究组总有效率为90.70%,显著高于对照组(69.39%,P0.05)。治疗前,两组血清BDNF、NGF、DA水平、MMSE和ADL评分比较差异无统计学意义(P0.05);治疗后,两组以上指标均较治疗前显著上升,且研究组以上指标明显高于对照组(P0.05)。两组不良反应总发生率比较差异无统计学意义(P0.05)。结论:盐酸美金刚联合银杏叶提取物可提高阿尔茨海默病的疗效,改善患者认知功能,可能与增加BDNF、NGF、DA的表达有关。     

Primary Biliary Acids Inhibit Hepatitis D Virus (HDV) Entry into Human Hepatoma Cells Expressing the Sodium-Taurocholate Cotransporting Polypeptide (NTCP)

BackgroundThe sodium-taurocholate cotransporting polypeptide (NTCP) is both a key bile acid (BA) transporter mediating uptake of BA into hepatocytes and an essential receptor for hepatitis B virus (HBV) and hepatitis D virus (HDV). In this study we aimed to characterize to what extent and through what mechanism BA affect HDV cell entry.MethodsHuH-7 cells stably expressing NTCP (HuH-7/NTCP) and primary human hepatocytes (PHH) were infected with in vitro generated HDV particles. Infectivity in the absence or presence of compounds was assessed using immunofluorescence staining for HDV antigen, standard 50% tissue culture infectious dose (TCID50) assays and quantitative PCR.ResultsAddition of primary conjugated and unconjugated BA resulted in a dose dependent reduction in the number of infected cells while secondary, tertiary and synthetic BA had a lesser effect. This effect was observed both in HuH-7/NTCP and in PHH. Other replication cycle steps such as replication and particle assembly and release were unaffected. Moreover, inhibitory BA competed with a fragment from the large HBV envelope protein for binding to NTCP-expressing cells. Conversely, the sodium/BA-cotransporter function of NTCP seemed not to be required for HDV infection since infection was similar in the presence or absence of a sodium gradient across the plasma membrane. When chenodeoxycolic acid (15 mg per kg body weight) was administered to three chronically HDV infected individuals over a period of up to 16 days there was no change in serum HDV RNA.ConclusionsPrimary BA inhibit NTCP-mediated HDV entry into hepatocytes suggesting that modulation of the BA pool may affect HDV infection of hepatocytes.     

美金刚联合普拉克索治疗帕金森患者临床疗效及对CysC、Hcy水平影响

目的:探究美金刚联合普拉克索在治疗帕金森患者中的临床疗效,并就治疗对患者胱抑素C(cystatin C,CysC)以及血同型半胱氨酸(homocysteine,Hcy)水平的影响。方法:选择2018年1月至2020年1月于我院接受治疗的98例帕金森患者,随机数字表法均分为两组(每组各49例),对照组单纯接受美金刚治疗,研究组在对照组基础上加用普拉克索进行治疗,对比两组治疗有效率,治疗前后CysC、Hcy水平,以及治疗前后简易智能精神状态检查量表(mini mental state examination,MMSE)以及帕金森病评定量表III(parkinson comprehensive rating scale,UPDRS III)评分,最后对两组患者治疗中不良反应发生率进行统计对比。结果:(1)研究组患者治疗有效率明显高于对照组患者(P0.05);(2)治疗前两组患者CysC、Hcy水平对比差异不具有统计学意义(P0.05),治疗后研究组患者CysC、Hcy水平低于对照组(P0.05);(3)治疗前两组患者MMSE及UPDRS III量表评分对比差异不具有统计学意义(P0.05),治疗后研究组患者MMSE得分高于对照组,UPDRS III量表评分低于对照组(P0.05);(4)两组治疗不良反应诸如胃肠道反应、嗜睡、体位性低血压等对比无差异(P0.05)。结论:美金刚联合普拉克索对帕金森具有较好的治疗效果,能够显著改善患者认知及运动功能,降低CysC、Hcy水平,同时治疗安全性较高。     

Relations between doses cumulated in bone marrow and dose delivery techniques during radiation therapy of cervical and endometrial cancer

PurposeTo compare normal tissue complication probability (NTCP) and average doses in the bone marrow (BM), obtained for five different radiotherapy delivery and planning strategies of cervical and endometrial cancer.Material/methods50 patients were taken to analysis. For each case, 3 different dose delivery techniques were used: 4-field, X15MV, 3DCRT; 7-field, X6MV, IMRT; and 2-arc, X6MV, VMAT. Two optimization scenarios were used for the IMRT and VMAT plans generation: with (+) and without (−) the inclusion of the BM as an optimized structure. Average doses and dose-volume histogram parameters for the PTV, BM, bladder, rectum, bowels and femoral heads were compared. In addition, the BM doses were analyzed with respect to the PTV and/or volume of the BM, and NTCP for the BM were computed.ResultsThe dose in PTV for evaluated plans was similar. The worst doses in organs at risk were obtained for 3DCRT. Using the BM during the optimization of IMRT and VMAT reduces an average dose in BM without increasing the doses in the bladder, rectum and bowels. Differences between doses in BM for IMRT(+) and VMAT(+) plans were similar while NTCP was lower for VMAT(+). A correlation between average dose in BM and the volume ratio of BM and PTV was found for each technique.ConclusionUsing the BM during the optimization of the IMRT and VMAT plans effectively reduces the dose in BM without increasing the dose in the bladder, rectum and bowels. The VMAT(+) plans were characterized by the lowest NTCP.     

Implementation of an Electronic Monitoring and Evaluation System for the Antiretroviral Treatment Programme in the Cape Winelands District,South Africa: A Qualitative Evaluation

BackgroundA pragmatic three-tiered approach to monitor the world’s largest antiretroviral treatment (ART) programme was adopted by the South African National Department of Health in 2010. With the rapid expansion of the programme, the limitations of the paper-based register (tier 1) were the catalyst for implementation of the stand-alone electronic register (tier 2), which offers simple digitisation of the paper-based register. This article engages with theory on implementation to identify and contextualise enabling and constraining factors for implementation of the electronic register, to describe experiences and use of the register, and to make recommendations for implementation in similar settings where standardisation of ART monitoring and evaluation has not been achieved.MethodsWe conducted a qualitative evaluation of the roll-out of the register. This comprised twenty in-depth interviews with a diverse sample of stakeholders at facility, sub-district, and district levels of the health system. Facility-level participants were selected across five sub-districts, including one facility per sub-district. Responses were coded and analysed using a thematic approach. An implementation science framework guided interpretation of the data.ConclusionIn this study we found that relative advantage of an intervention and stakeholder engagement are critical to implementation. We suggest that without these aspects of implementation, formative and summative outcomes of implementation at both the adoption and coalface stages of implementation would be negatively affected.     

Critical success factors for implementation of an incident learning system in radiation oncology department

AimThe aim of this study was to analyze critical success factors (CSFs) for implementation of an incident learning system (ILS) in a radiation oncology department (ROD) and evaluate the perception of the staff members along this process.BackgroundImplementing an ILS is a way to leverage learning from incidents and is a tool for improving patient safety, consisting of a cycle of reporting and analyzing events as well as taking preventive actions. ILS implementation is challenging, requiring specific resources and cultural changes.Materials and methodsAn ILS was designed and implemented based on the CSF identified in the literature review. Before starting the ILS implementation, a structured survey was applied to assess dimensions of patient safety culture. After the period of implementation (7 months), the survey was applied again and compared with the initial assessment, and interviews were performed with staff members to evaluate the overall satisfaction with ILS and CSFs.ResultsStatistically significant improvements were observed in 5 dimensions (12 totals) of the safety culture survey, considering time points before and after the ILS implementation. According to interviewees, “Facilitating committee”, “Efficient data collection”, “Focus on improvement”, “Just culture” and “Feedback to users” were the most relevant CSFs.ConclusionsThe ILS designed and implemented at ROD was perceived as an important tool to support quality and safety initiatives, promoting the improvement in safety culture. The ILS implementation critical success factors were identified and have shown good agreement between the results of the literature and the users' practical perception.     

Multi-input deep learning approach for Cardiovascular Disease diagnosis using Myocardial Perfusion Imaging and clinical data

PurposeAccurate detection and treatment of Coronary Artery Disease is mainly based on invasive Coronary Angiography, which could be avoided provided that a robust, non-invasive detection methodology emerged. Despite the progress of computational systems, this remains a challenging issue. The present research investigates Machine Learning and Deep Learning methods in competing with the medical experts' diagnostic yield. Although the highly accurate detection of Coronary Artery Disease, even from the experts, is presently implausible, developing Artificial Intelligence models to compete with the human eye and expertise is the first step towards a state-of-the-art Computer-Aided Diagnostic system.MethodsA set of 566 patient samples is analysed. The dataset contains Polar Maps derived from scintigraphic Myocardial Perfusion Imaging studies, clinical data, and Coronary Angiography results. The latter is considered as reference standard. For the classification of the medical images, the InceptionV3 Convolutional Neural Network is employed, while, for the categorical and continuous features, Neural Networks and Random Forest classifier are proposed.ResultsThe research suggests that an optimal strategy competing with the medical expert's accuracy involves a hybrid multi-input network composed of InceptionV3 and a Random Forest. This method matches the expert's accuracy, which is 79.15% in the particular dataset.ConclusionImage classification using deep learning methods can cooperate with clinical data classification methods to enhance the robustness of the predicting model, aiming to compete with the medical expert's ability to identify Coronary Artery Disease subjects, from a large scale patient dataset.     

基于BP神经网络和RBF神经网络预测老年痴呆症疾病进展的对比研究

目的:比较反向传播算法(BP)神经网络和径向基函数(RBF)神经网络预测老年痴呆症疾病进展的效果。方法:以老年痴呆症随访数据为研究对象,以性别、年龄、受教育程度、有无高血压、有无高胆固醇、有无心脏病、有无中风史、有无家族史8个指标作为输入变量,以五年随访的MMSE差值为输出变量,构建基于BP神经网络和RBF神经网络的老年痴呆症疾病进展预测模型。结果:与BP神经网络模型相比,RBF神经网络预测的结果更好,能够有效地预测老年痴呆症疾病进展。结论:神经网络模型将老年痴呆症疾病进展预测问题转化为随访数据中相关测量指标与MMSE差值的非线性问题,为复杂的老年痴呆症疾病进展预测提供了新思路。     

Image guidance in clinical practice – Influence of positioning inaccuracy on the dose distribution for prostate cancer

BackgroundIn order to consider potential positioning errors there are different recipes for safety-margins for CTV-to-PTV expansion. The aim of this study is to simulate the effect of positioning inaccuracy with clinically realistic patient treatment plans.MethodsFor a collective of 40 prostate patients, the isocenter was shifted back appropriately to the applied table shifts after positioning verification, simulating that no positioning correction had been performed and the treatment plans were recalculated. All the treatment fractions with the appropriate isocenter-shifts were added to yield a new plan considering two scenarios:

• 1)Extreme scenario: summation of only shifted plans.
• 2)Realistic scenario: consideration of the original treatment plan for the fractions with verification imaging.

Afterwards all plans were analysed and compared with each other regarding target coverage, sparing of organs at risk (OAR) and normal tissue complication probability (NTCP).ResultsDose distributions and especially DVH show a deterioration of the target-coverage caused by the positioning inaccuracy. Deviations in dose at a single point can reach values of over 10 Gy. In single cases minimum plan agreement only achieved 66% pass within 3% local dose for the realistic case. Organs at risk and NTCP analysis result in a slightly better sparing of the rectum. Measures of quality like homogeneity and conformity differ just minimally regarding the different scenarios.ConclusionPTV-coverage suffers markedly by the positioning uncertainties, the shifted plans are in large parts clinically not acceptable. Surprisingly sparing of the OAR is not negatively affected by potential positioning errors for this prostate collective.