The development of Monte Carlo techniques for the verification of radiotherapy treatments

Introduction: Recent advances in the planning and delivery of radiotherapy treatments have resulted in improvements in the accuracy and precision with which therapeutic radiation can be administered. As the complexity of the treatments increases it becomes more difficult to predict the dose distribution in the patient accurately. Monte Carlo (MC) methods have the potential to improve the accuracy of the dose calculations and are increasingly being recognised as the ‘gold standard’ for predicting dose deposition in the patient [1]. This project has three main aims: 1. To develop tools that enable the transfer of treatment plan information from the treatment planning system (TPS) to a MC dose calculation engine. 2. To develop tools for comparing the 3D dose distributions calculated by the TPS and the MC dose engine. 3. To investigate the radiobiological significance of any errors between the TPS patient dose distribution and the MC dose distribution in terms of Tumour Control Probability (TCP) and Normal Tissue Complication Probabilities (NTCP). The work presented here addresses the first two aims. Methods: (1a) Plan Importing: A database of commissioned accelerator models (Elekta Precise and Varian 2100CD) has been developed for treatment simulations in the MC system (EGSnrc/BEAMnrc). Beam descriptions can be exported from the TPS using the widespread DICOM framework, and the resultant files are parsed with the assistance of a software library (PixelMed Java DICOM Toolkit). The information in these files (such as the monitor units, the jaw positions and gantry orientation) is used to construct a plan-specific accelerator model which allows an accurate simulation of the patient treatment field. (1b) Dose Simulation: The calculation of a dose distribution requires patient CT images which are prepared for the MC simulation using a tool (CTCREATE) packaged with the system. Beam simulation results are converted to absolute dose per- MU using calibration factors recorded during the commissioning process and treatment simulation. These distributions are combined according to the MU meter settings stored in the exported plan to produce an accurate description of the prescribed dose to the patient. (2) Dose Comparison: TPS dose calculations can be obtained using either a DICOM export or by direct retrieval of binary dose files from the file system. Dose difference, gamma evaluation and normalised dose difference algorithms [2] were employed for the comparison of the TPS dose distribution and the MC dose distribution. These implementations are spatial resolution independent and able to interpolate for comparisons. Results and Discussion: The tools successfully produced Monte Carlo input files for a variety of plans exported from the Eclipse (Varian Medical Systems) and Pinnacle (Philips Medical Systems) planning systems: ranging in complexity from a single uniform square field to a five-field step and shoot IMRT treatment. The simulation of collimated beams has been verified geometrically, and validation of dose distributions in a simple body phantom (QUASAR) will follow. The developed dose comparison algorithms have also been tested with controlled dose distribution changes. Conclusion: The capability of the developed code to independently process treatment plans has been demonstrated. A number of limitations exist: only static fields are currently supported (dynamic wedges and dynamic IMRT will require further development), and the process has not been tested for planning systems other than Eclipse and Pinnacle. The tools will be used to independently assess the accuracy of the current treatment planning system dose calculation algorithms for complex treatment deliveries such as IMRT in treatment sites where patient inhomogeneities are expected to be significant. Acknowledgements: Computational resources and services used in this work were provided by the HPC and Research Support Group, Queensland University of Technology, Brisbane, Australia. Pinnacle dose parsing made possible with the help of Paul Reich, North Coast Cancer Institute, North Coast, New South Wales.

Investigation of megavoltage beams and detectors for electronic portal imaging

Accuracy of dose delivery in external beam radiotherapy is usually verified with electronic portal imaging (EPI) in which the treatment beam is used to check the positioning of the patient. However the resulting megavoltage x-ray images suffer from poor quality. The image quality can be improved by developing a special operating mode in the linear accelerator. The existing treatment beam is modified such that it produces enough low-energy photons for imaging. In this work the problem of optimizing the beam/detector combination to achieve optimal electronic portal image quality is addressed. The linac used for this study was modified to produce two experimental photon beams. These beams, named Al6 and Al10, were non-flat and were produced by 4MeV electrons hitting aluminum targets, 6 and 10mm thick respectively. The images produced by a conventional EPI system (6MV treatment beam and camera-based EPID with a Cu plate & Gd2O2S screen ) were compared with the images produced by the experimental beams and various screens with the same camera). The contrast of 0.8cm bone equivalent material in 5 cm water increased from 1.5% for the conventional system to 11% for the combination of Al6 beam with a 200mg/cm2 Gd2O2S screen. The signal-to-noise ratio calculated for 1cGy flood field images increased by about a factor of two for the same EPI systems. The spatial resolution of the two imaging systems was comparable. This work demonstrates that significant improvements in portal image contrast can be obtained by simultaneous optimization of the linac spectrum and EPI detector.

Nutritional status of Ugandan women living with HIV/AIDS : anthropometry and body composition assessment

Human immunodeficiency virus (HIV) that leads to acquired immune deficiency syndrome (AIDs) reduces immune function, resulting in opportunistic infections and later death. Use of antiretroviral therapy (ART) increases chances of survival, however, with some concerns regarding fat re-distribution (lipodystrophy) which may encompass subcutaneous fat loss (lipoatrophy) and/or fat accumulation (lipohypertrophy), in the same individual. This problem has been linked to Antiretroviral drugs (ARVs), majorly, in the class of protease inhibitors (PIs), in addition to older age and being female. An additional concern is that the problem exists together with the metabolic syndrome, even when nutritional status/ body composition, and lipodystrophy/metabolic syndrome are unclear in Uganda where the use of ARVs is on the increase. In line with the literature, the overall aim of the study was to assess physical characteristics of HIV-infected patients using a comprehensive anthropometric protocol and to predict body composition based on these measurements and other standardised techniques. The other aim was to establish the existence of lipodystrophy, the metabolic syndrome, andassociated risk factors. Thus, three studies were conducted on 211 (88 ART-naïve) HIV-infected, 15-49 year-old women, using a cross-sectional approach, together with a qualitative study of secondary information on patient HIV and medication status. In addition, face-to-face interviews were used to extract information concerning morphological experiences and life style. The study revealed that participants were on average 34.1±7.65 years old, had lived 4.63±4.78 years with HIV infection and had spent 2.8±1.9 years receiving ARVs. Only 8.1% of participants were receiving PIs and 26% of those receiving ART had ever changed drug regimen, 15.5% of whom changed drugs due to lipodystrophy. Study 1 hypothesised that the mean nutritional status and predicted percent body fat values of study participants was within acceptable ranges; different for participants receiving ARVs and the HIV-infected ART-naïve participants and that percent body fat estimated by anthropometric measures (BMI and skinfold thickness) and the BIA technique was not different from that predicted by the deuterium oxide dilution technique. Using the Body Mass Index (BMI), 7.1% of patients were underweight (<18.5 kg/m2) and 46.4% were overweight/obese (≥25.0 kg/m2). Based on waist circumference (WC), approximately 40% of the cohort was characterized as centrally obese. Moreover, the deuterium dilution technique showed that there was no between-group difference in the total body water (TBW), fat mass (FM) and fat-free mass (FFM). However, the technique was the only approach to predict a between-group difference in percent body fat (p = .045), but, with a very small effect (0.021). Older age (β = 0.430, se = 0.089, p = .000), time spent receiving ARVs (β = 0.972, se = 0.089, p = .006), time with the infection (β = 0.551, se = 0.089, p = .000) and receiving ARVs (β = 2.940, se = 1.441, p = .043) were independently associated with percent body fat. Older age was the greatest single predictor of body fat. Furthermore, BMI gave better information than weight alone could; in that, mean percentage body fat per unit BMI (N = 192) was significantly higher in patients receiving treatment (1.11±0.31) vs. the exposed group (0.99±0.38, p = .025). For the assessment of obesity, percent fat measures did not greatly alter the accuracy of BMI as a measure for classifying individuals into the broad categories of underweight, normal and overweight. Briefly, Study 1 revealed that there were more overweight/obese participants than in the general Ugandan population, the problem was associated with ART status and that BMI broader classification categories were maintained when compared with the gold standard technique. Study 2 hypothesized that the presence of lipodystrophy in participants receiving ARVs was not different from that of HIV-infected ART-naïve participants. Results showed that 112 (53.1%) patients had experienced at least one morphological alteration including lipohypertrophy (7.6%), lipoatrophy (10.9%), and mixed alterations (34.6%). The majority of these subjects (90%) were receiving ARVs; in fact, all patients receiving PIs reported lipodystrophy. Period spent receiving ARVs (t209 = 6.739, p = .000), being on ART (χ2 = 94.482, p = .000), receiving PIs (Fisher’s exact χ2 = 113.591, p = .000), recent T4 count (CD4 counts) (t207 = 3.694, p = .000), time with HIV (t125 = 1.915, p = .045), as well as older age (t209 = 2.013, p = .045) were independently associated with lipodystrophy. Receiving ARVs was the greatest predictor of lipodystrophy (p = .000). In other analysis, aside from skinfolds at the subscapular (p = .004), there were no differences with the rest of the skinfold sites and the circumferences between participants with lipodystrophy and those without the problem. Similarly, there was no difference in Waist: Hip ratio (WHR) (p = .186) and Waist: Height ratio (WHtR) (p = .257) among participants with lipodystrophy and those without the problem. Further examination showed that none of the 4.1% patients receiving stavudine (d4T) did experience lipoatrophy. However, 17.9% of patients receiving EFV, a non-nucleoside reverse transcriptase inhibitor (NNRTI) had lipoatrophy. Study 2 findings showed that presence of lipodystrophy in participants receiving ARVs was in fact far higher than that of HIV-infected ART-naïve participants. A final hypothesis was that the prevalence of the metabolic syndrome in participants receiving ARVs was not different from that of HIV-infected ART-naïve participants. Moreover, data showed that many patients (69.2%) lived with at least one feature of the metabolic syndrome based on International Diabetic Federation (IDF, 2006) definition. However, there was no single anthropometric predictor of components of the syndrome, thus, the best anthropometric predictor varied as the component varied. The metabolic syndrome was diagnosed in 15.2% of the subjects, lower than commonly reported in this population, and was similar between the medicated and the exposed groups (χ 21 = 0.018, p = .893). Moreover, the syndrome was associated with older age (p = .031) and percent body fat (p = .012). In addition, participants with the syndrome were heavier according to BMI (p = .000), larger at the waist (p = .000) and abdomen (p = .000), and were at central obesity risk even when hip circumference (p = .000) and height (p = .000) were accounted for. In spite of those associations, results showed that the period with disease (p = .13), CD4 counts (p = .836), receiving ART (p = .442) or PIs (p = .678) were not associated with the metabolic syndrome. While the prevalence of the syndrome was highest amongst the older, larger and fatter participants, WC was the best predictor of the metabolic syndrome (p = .001). Another novel finding was that participants with the metabolic syndrome had greater arm muscle circumference (AMC) (p = .000) and arm muscle area (AMA) (p = .000), but the former was most influential. Accordingly, the easiest and cheapest indicator to assess risk in this study sample was WC should routine laboratory services not be feasible. In addition, the final study illustrated that the prevalence of the metabolic syndrome in participants receiving ARVs was not different from that of HIV-infected ART-naïve participants.

Segmentation of cone-beam CT using a hidden Markov random field with informative priors

Purpose: Flat-detector, cone-beam computed tomography (CBCT) has enormous potential to improve the accuracy of treatment delivery in image-guided radiotherapy (IGRT). To assist radiotherapists in interpreting these images, we use a Bayesian statistical model to label each voxel according to its tissue type. Methods: The rich sources of prior information in IGRT are incorporated into a hidden Markov random field (MRF) model of the 3D image lattice. Tissue densities in the reference CT scan are estimated using inverse regression and then rescaled to approximate the corresponding CBCT intensity values. The treatment planning contours are combined with published studies of physiological variability to produce a spatial prior distribution for changes in the size, shape and position of the tumour volume and organs at risk (OAR). The voxel labels are estimated using the iterated conditional modes (ICM) algorithm. Results: The accuracy of the method has been evaluated using 27 CBCT scans of an electron density phantom (CIRS, Inc. model 062). The mean voxel-wise misclassification rate was 6.2%, with Dice similarity coefficient of 0.73 for liver, muscle, breast and adipose tissue. Conclusions: By incorporating prior information, we are able to successfully segment CBCT images. This could be a viable approach for automated, online image analysis in radiotherapy.

Sampling stored-product insect pests : a comparison of four statistical sampling models for probability of pest detection

Background: Developing sampling strategies to target biological pests such as insects in stored grain is inherently difficult owing to species biology and behavioural characteristics. The design of robust sampling programmes should be based on an underlying statistical distribution that is sufficiently flexible to capture variations in the spatial distribution of the target species. Results: Comparisons are made of the accuracy of four probability-of-detection sampling models - the negative binomial model,1 the Poisson model,1 the double logarithmic model2 and the compound model3 - for detection of insects over a broad range of insect densities. Although the double log and negative binomial models performed well under specific conditions, it is shown that, of the four models examined, the compound model performed the best over a broad range of insect spatial distributions and densities. In particular, this model predicted well the number of samples required when insect density was high and clumped within experimental storages. Conclusions: This paper reinforces the need for effective sampling programs designed to detect insects over a broad range of spatial distributions. The compound model is robust over a broad range of insect densities and leads to substantial improvement in detection probabilities within highly variable systems such as grain storage.

Detection of a novel mutation in the CACNA1A gene

Familial hemiplegic migraine (FHM) is a rare autosomal dominant subtype of migraine with aura. It is divided into three subtypes FHM1, FHM2 and FHM3, which are caused by mutations in the CACNA1A, ATP1A2 and SCN1A genes respectively. As part of a regular diagnostic service, we investigated 168 patients with FHM symptoms. Samples were tested for mutations contained within the CACNA1A gene. Some tested samples (4.43%) showed an FHM1 mutation, with five of the mutations found in exon 5, one mutation in exon 16 and one in exon 17. Four polymorphisms were also detected, one of which occurred in a large percentage of samples (14.88%). The exon 16 2094G>A polymorphism, however, has been found to occur in healthy Caucasian control populations up to a frequency of 16% and is not considered to be significantly associated with FHM. A finding of significance, found in a single patient, was the detection of a novel mutation in exon 5 that results in a P225H change. The affected individual was an 8-year-old female. The exact phenotypic effect of this mutation is unknown, and further studies are needed to understand the pathophysiology of this mutation in FHM1. New information will allow for diagnostic procedures to be constantly updated, thus improving accuracy of diagnosis. It is possible that new information will also aid the development of new therapeutic agents for the treatment of FHM.

Understanding requirements of novel healthcare information systems for management of advanced prostate cancer

Effective management of chronic diseases is a global health priority. A healthcare information system offers opportunities to address challenges of chronic disease management. However, the requirements of health information systems are often not well understood. The accuracy of requirements has a direct impact on the successful design and implementation of a health information system. Our research describes methods used to understand the requirements of health information systems for advanced prostate cancer management. The research conducted a survey to identify heterogeneous sources of clinical records. Our research showed that the General Practitioner was the common source of patient's clinical records (41%) followed by the Urologist (14%) and other clinicians (14%). Our research describes a method to identify diverse data sources and proposes a novel patient journey browser prototype that integrates disparate data sources.

Travel time prediction on signalised urban arterials by applying SARIMA modelling on Bluetooth data

Travel time prediction has long been the topic of transportation research. But most relevant prediction models in the literature are limited to motorways. Travel time prediction on arterial networks is challenging due to involving traffic signals and significant variability of individual vehicle travel time. The limited availability of traffic data from arterial networks makes travel time prediction even more challenging. Recently, there has been significant interest of exploiting Bluetooth data for travel time estimation. This research analysed the real travel time data collected by the Brisbane City Council using the Bluetooth technology on arterials. Databases, including experienced average daily travel time are created and classified for approximately 8 months. Thereafter, based on data characteristics, Seasonal Auto Regressive Integrated Moving Average (SARIMA) modelling is applied on the database for short-term travel time prediction. The SARMIA model not only takes the previous continuous lags into account, but also uses the values from the same time of previous days for travel time prediction. This is carried out by defining a seasonality coefficient which improves the accuracy of travel time prediction in linear models. The accuracy, robustness and transferability of the model are evaluated through comparing the real and predicted values on three sites within Brisbane network. The results contain the detailed validation for different prediction horizons (5 min to 90 minutes). The model performance is evaluated mainly on congested periods and compared to the naive technique of considering the historical average.

Pre-crash traffic flow trend analysis on motorways

Crashes on motorway contribute to a significant proportion (40-50%) of non-recurrent motorway congestions. Hence reduce crashes will help address congestion issues (Meyer, 2008). Crash likelihood estimation studies commonly focus on traffic conditions in a Short time window around the time of crash while longer-term pre-crash traffic flow trends are neglected. In this paper we will show, through data mining techniques, that a relationship between pre-crash traffic flow patterns and crash occurrence on motorways exists, and that this knowledge has the potential to improve the accuracy of existing models and opens the path for new development approaches. The data for the analysis was extracted from records collected between 2007 and 2009 on the Shibuya and Shinjuku lines of the Tokyo Metropolitan Expressway in Japan. The dataset includes a total of 824 rear-end and sideswipe crashes that have been matched with traffic flow data of one hour prior to the crash using an incident detection algorithm. Traffic flow trends (traffic speed/occupancy time series) revealed that crashes could be clustered with regards of the dominant traffic flow pattern prior to the crash. Using the k-means clustering method allowed the crashes to be clustered based on their flow trends rather than their distance. Four major trends have been found in the clustering results. Based on these findings, crash likelihood estimation algorithms can be fine-tuned based on the monitored traffic flow conditions with a sliding window of 60 minutes to increase accuracy of the results and minimize false alarms.

Pre-crash and non-crash traffic flow trends analysis on motorways

Crashes that occur on motorways contribute to a significant proportion (40-50%) of non-recurrent motorway congestions. Hence, reducing the frequency of crashes assists in addressing congestion issues (Meyer, 2008). Crash likelihood estimation studies commonly focus on traffic conditions in a short time window around the time of a crash while longer-term pre-crash traffic flow trends are neglected. In this paper we will show, through data mining techniques that a relationship between pre-crash traffic flow patterns and crash occurrence on motorways exists. We will compare them with normal traffic trends and show this knowledge has the potential to improve the accuracy of existing models and opens the path for new development approaches. The data for the analysis was extracted from records collected between 2007 and 2009 on the Shibuya and Shinjuku lines of the Tokyo Metropolitan Expressway in Japan. The dataset includes a total of 824 rear-end and sideswipe crashes that have been matched with crashes corresponding to traffic flow data using an incident detection algorithm. Traffic trends (traffic speed time series) revealed that crashes can be clustered with regards to the dominant traffic patterns prior to the crash. Using the K-Means clustering method with Euclidean distance function allowed the crashes to be clustered. Then, normal situation data was extracted based on the time distribution of crashes and were clustered to compare with the “high risk” clusters. Five major trends have been found in the clustering results for both high risk and normal conditions. The study discovered traffic regimes had differences in the speed trends. Based on these findings, crash likelihood estimation models can be fine-tuned based on the monitored traffic conditions with a sliding window of 30 minutes to increase accuracy of the results and minimize false alarms.

Segmentation of cone-beam CT using a hidden Markov random field with informative priors

Cone-beam computed tomography (CBCT) has enormous potential to improve the accuracy of treatment delivery in image-guided radiotherapy (IGRT). To assist radiotherapists in interpreting these images, we use a Bayesian statistical model to label each voxel according to its tissue type. The rich sources of prior information in IGRT are incorporated into a hidden Markov random field model of the 3D image lattice. Tissue densities in the reference CT scan are estimated using inverse regression and then rescaled to approximate the corresponding CBCT intensity values. The treatment planning contours are combined with published studies of physiological variability to produce a spatial prior distribution for changes in the size, shape and position of the tumour volume and organs at risk. The voxel labels are estimated using iterated conditional modes. The accuracy of the method has been evaluated using 27 CBCT scans of an electron density phantom. The mean voxel-wise misclassification rate was 6.2\%, with Dice similarity coefficient of 0.73 for liver, muscle, breast and adipose tissue. By incorporating prior information, we are able to successfully segment CBCT images. This could be a viable approach for automated, online image analysis in radiotherapy.

A methodology to develop an urban transport disadvantage framework : the case of Brisbane, Australia

Most individuals travel in order to participate in a network of activities which are important for attaining a good standard of living. Because such activities are commonly widely dispersed and not located locally, regular access to a vehicle is important to avoid exclusion. However, planning transport system provisions that can engage members of society in an acceptable degree of activity participation remains a great challenge. The main challenges in most cities of the world are due to significant population growth and rapid urbanisation which produces increased demand for transport. Keeping pace with these challenges in most urban areas is difficult due to the widening gap between supply and demand for transport systems which places the urban population at a transport disadvantage. The key element in mitigating the issue of urban transport disadvantage is to accurately identify the urban transport disadvantaged. Although wide-ranging variables and multi-dimensional methods have been used to identify this group, variables are commonly selected using ad-hoc techniques and unsound methods. This poses questions of whether the current variables used are accurately linked with urban transport disadvantage, and the effectiveness of the current policies. To fill these gaps, the research conducted for this thesis develops an operational urban transport disadvantage framework (UTDAF) based on key statistical urban transport disadvantage variables to accurately identify the urban transport disadvantaged. The thesis develops a methodology based on qualitative and quantitative statistical approaches to develop an urban transport disadvantage framework designed to accurately identify urban transport disadvantage. The reliability and the applicability of the methodology developed is the prime concern rather than the accuracy of the estimations. Relevant concepts that impact on urban transport disadvantage identification and measurement and a wide range of urban transport disadvantage variables were identified through a review of the existing literature. Based on the reviews, a conceptual urban transport disadvantage framework was developed based on the causal theory. Variables identified during the literature review were selected and consolidated based on the recommendations of international and local experts during the Delphi study. Following the literature review, the conceptual urban transport disadvantage framework was statistically assessed to identify key variables. Using the statistical outputs, the key variables were weighted and aggregated to form the UTDAF. Before the variable's weights were finalised, they were adjusted based on results of correlation analysis between elements forming the framework to improve the framework's accuracy. The UTDAF was then applied to three contextual conditions to determine the framework's effectiveness in identifying urban transport disadvantage. The development of the framework is likely to be a robust application measure for policy makers to justify infrastructure investments and to generate awareness about the issue of urban transport disadvantage.

The consistency of crash involvement recall across time

Introduction Road safety researchers rely heavily on self-report data to explore the aetiology of crash risk. However, researchers consistently acknowledge a range of limitations associated with this methodological approach (e.g., self-report bias), which has been hypothesised to reduce the predictive efficacy of scales. Although well researched in other areas, one important factor often neglected in road safety studies is the fallibility of human memory. Given accurate recall is a key assumption in many studies, the validity and consistency of self-report data warrants investigation. The aim of the current study was to examine the consistency of self-report data of crash history and details of the most recent reported crash on two separate occasions. Materials & Method A repeated measures design was utilised to examine the self-reported crash involvement history of 214 general motorists over a two month period. Results A number of interesting discrepancies were noted in relation to number of lifetime crashes reported by the participants and the descriptions of their most recent crash across the two occasions. Of the 214 participants who reported having been involved in a crash, 35 (22.3%) reported a lower number of lifetime crashes as Time 2, than at Time 1. Of the 88 drivers who reported no change in number of lifetime crashes, 10 (11.4%) described a different most recent crash. Additionally, of the 34 reporting an increase in the number of lifetime crashes, 29 (85.3%) of these described the same crash on both occasions. Assessed as a whole, at least 47.1% of participants made a confirmed mistake at Time 1 or Time 2. Conclusions These results raise some doubt in regard to the accuracy of memory recall across time. Given that self-reported crash involvement is the predominant dependent variable used in the majority of road safety research, this issue warrants further investigation. Replication of the study with a larger sample size that includes multiple recall periods would enhance understanding into the significance of this issue for road safety methodology.

Review of current test methods for screwed connections

The pull-through/local dimpling failure strength of screwed connections is very important in the design of profiled steel cladding systems to help them resist storms and hurricanes. The current American and European provisions recommend four different test methods for the screwed connections in tension, but the accuracy of these methods in determining the connection strength is not known. It is unlikely that the four test methods are equivalent in all cases and thus it is necessary to reduce the number of methods recommended. This paper presents a review of these test methods based on some laboratory tests on crest- and valley-fixed claddings and then recommends alternative tests methods that reproduce the real behavior of the connections, including the bending and membrane deformations of the cladding around the screw fasteners and the tension load in the fastener.

Traffic safety risks trends and patterns analysis on motorways

Crashes that occur on motorways contribute to a significant proportion (40-50%) of non-recurrent motorway congestion. Hence, reducing the frequency of crashes assist in addressing congestion issues (Meyer, 2008). Analysing traffic conditions and discovering risky traffic trends and patterns are essential basics in crash likelihood estimations studies and still require more attention and investigation. In this paper we will show, through data mining techniques, that there is a relationship between pre-crash traffic flow patterns and crash occurrence on motorways, compare them with normal traffic trends, and that this knowledge has the potentiality to improve the accuracy of existing crash likelihood estimation models, and opens the path for new development approaches. The data for the analysis was extracted from records collected between 2007 and 2009 on the Shibuya and Shinjuku lines of the Tokyo Metropolitan Expressway in Japan. The dataset includes a total of 824 rear-end and sideswipe crashes that have been matched with crashes corresponding traffic flow data using an incident detection algorithm. Traffic trends (traffic speed time series) revealed that crashes can be clustered with regards to the dominant traffic patterns prior to the crash occurrence. K-Means clustering algorithm applied to determine dominant pre-crash traffic patterns. In the first phase of this research, traffic regimes identified by analysing crashes and normal traffic situations using half an hour speed in upstream locations of crashes. Then, the second phase investigated the different combination of speed risk indicators to distinguish crashes from normal traffic situations more precisely. Five major trends have been found in the first phase of this paper for both high risk and normal conditions. The study discovered traffic regimes had differences in the speed trends. Moreover, the second phase explains that spatiotemporal difference of speed is a better risk indicator among different combinations of speed related risk indicators. Based on these findings, crash likelihood estimation models can be fine-tuned to increase accuracy of estimations and minimize false alarms.