Browsing by Subject "segmentation"

Sort by: Order: Results:

Now showing items 1-5 of 5
  • Akmal, Jan Sher; Salmi, Mika; Hemming, Björn; Teir, Linus; Suomalainen, Anni; Kortesniemi, Mika; Partanen, Jouni; Lassila, Antti (2020)
    Featured Application Accuracy of additively manufactured implants for clinical surgery. Abstract In craniomaxillofacial surgical procedures, an emerging practice adopts the preoperative virtual planning that uses medical imaging (computed tomography), 3D thresholding (segmentation), 3D modeling (digital design), and additive manufacturing (3D printing) for the procurement of an end-use implant. The objective of this case study was to evaluate the cumulative spatial inaccuracies arising from each step of the process chain when various computed tomography protocols and thresholding values were independently changed. A custom-made quality assurance instrument (Phantom) was used to evaluate the medical imaging error. A sus domesticus (domestic pig) head was analyzed to determine the 3D thresholding error. The 3D modeling error was estimated from the computer-aided design software. Finally, the end-use implant was used to evaluate the additive manufacturing error. The results were verified using accurate measurement instruments and techniques. A worst-case cumulative error of 1.7 mm (3.0%) was estimated for one boundary condition and 2.3 mm (4.1%) for two boundary conditions considering the maximum length (56.9 mm) of the end-use implant. Uncertainty from the clinical imaging to the end-use implant was 0.8 mm (1.4%). This study helps practitioners establish and corroborate surgical practices that are within the bounds of an appropriate accuracy for clinical treatment and restoration.
  • Heinonen, Kristina; Mickelsson, Jakob; Strandvik, Tore (Svenska handelshögskolan, 2010)
    Working Papers
    All companies have a portfolio of customer relationships. From a managerial standpoint the value of these customer relationships is a key issue. The aim of the paper is to introduce a conceptual framework for customers’ energy towards a service provider. Customer energy is defined as the cognitive, affective and behavioural effort a customer puts into the purchase of an offering. It is based on two dimensions: life theme involvement and relationship commitment. Data from a survey study of 425 customers of an online gambling site was combined with data about their individual purchases and activity. Analysis showed that involvement and commitment influence both customer behaviour and attitudes. Customer involvement was found to be strongly related to overall spending within a consumption area, whereas relationship commitment is a better predictor of the amount of money spent at a particular company. Dividing the customers into four different involvement / commitment segments revealed differences in churn rates, word-of-mouth, brand attitude, switching propensity and the use of the service for socializing. The framework provides a tool for customer management by revealing differences in fundamental drivers of customer behaviour resulting in completely new customer portfolios. Knowledge of customer energy allows companies to manage their communication and offering development better and provides insight into the risk of losing a customer.
  • Torkko, Jussi (Helsingin yliopisto, 2021)
    Urban greenery is vital to the people in our increasingly urbanizing societies. It is diverse in nature and provides numerous life improving qualities. Traditionally urban greenery has been assessed with a top-down view through the sensors of aerial vehicles and satellites. This does not equate on what is experienced down at the human level. An alternative viewpoint has emerged, with the introduction of a more human-scale viewpoint. To quantify this human-scale greenery, novel and disparate approaches have been developed. However, there is little knowledge on how these modelling methods and indices manage to capture the greenery we truly experience on the ground level. This thesis is an undertaking to better understand what the greenery experienced by people on the ground level, termed humanscale greenery (HSG), means. The goal was to grasp how the various modelling methods, indices and datasets can be best used to capture this phenomenon. Simultaneously, the study tries to better comprehend how different people experience greenery. To achieve this, human-scale greenery values were collected via interviews at randomly selected study sites across Helsinki. These values were then compared to modelled values at the same sites. The methods and indices tested include modern approaches developed specifically for HSG and traditional greenery assessment methods. Along the greenery values, sociodemographic variables were collected in the interviews and compared to each other in relation to HSG values. The modelled values were on average smaller than HSG values. All methods indicated very strong or strong linear relationships with human-scale greenery. NDVI and semantic segmentation Green View Index (GVI) had the strongest relationships and least deviation. Land use (LU) and color based GVI had the highest error deviations from HSG. The sociodemographic assessment showed hints that age might affect the amount of experienced greenery, but this is uncertain. With a random sampling of interviewees, 25–34-year-olds and less nature visiting people were more common at sites with low HSG. Based on the results obtained here, many different types of novel methods are suitable for modelling HSG with strong linear relationships. However, also traditional greenery assessment methods performed well. It is difficult to curtail the experience of greenery into a single approach. A solution could possibly be obtained via the combination of methods. The results also advocate the usage of machine learning methods for greenery image segmentation. These cannot be applied everywhere due to data coverage problems, but alternative methods can also be used to fill in gaps. The significance of age on the experience of greenery needs further research. Because urban greenery’s benefits are known, attention should also be given onto how different kinds of people are able to experience it. In the future we should also discuss the meaningfulness of assessing absolute greenery in comparison to the types and parts of greenery.
  • Rasse, Tobias M.; Hollandi, Reka; Horvath, Peter (2020)
    Various pre-trained deep learning models for the segmentation of bioimages have been made available as developer-to-end-user solutions. They are optimized for ease of use and usually require neither knowledge of machine learning nor coding skills. However, individually testing these tools is tedious and success is uncertain. Here, we present the Open Segmentation Framework (OpSeF), a Python framework for deep learning-based instance segmentation. OpSeF aims at facilitating the collaboration of biomedical users with experienced image analysts. It builds on the analysts' knowledge in Python, machine learning, and workflow design to solve complex analysis tasks at any scale in a reproducible, well-documented way. OpSeF defines standard inputs and outputs, thereby facilitating modular workflow design and interoperability with other software. Users play an important role in problem definition, quality control, and manual refinement of results. OpSeF semi-automates preprocessing, convolutional neural network (CNN)-based segmentation in 2D or 3D, and postprocessing. It facilitates benchmarking of multiple models in parallel. OpSeF streamlines the optimization of parameters for pre- and postprocessing such, that an available model may frequently be used without retraining. Even if sufficiently good results are not achievable with this approach, intermediate results can inform the analysts in the selection of the most promising CNN-architecture in which the biomedical user might invest the effort of manually labeling training data. We provide Jupyter notebooks that document sample workflows based on various image collections. Analysts may find these notebooks useful to illustrate common segmentation challenges, as they prepare the advanced user for gradually taking over some of their tasks and completing their projects independently. The notebooks may also be used to explore the analysis options available within OpSeF in an interactive way and to document and share final workflows. Currently, three mechanistically distinct CNN-based segmentation methods, the U-Net implementation used in Cellprofiler 3.0, StarDist, and Cellpose have been integrated within OpSeF. The addition of new networks requires little; the addition of new models requires no coding skills. Thus, OpSeF might soon become both an interactive model repository, in which pre-trained models might be shared, evaluated, and reused with ease.
  • Ahlgren, Niklas; Sjöö, Boo (Svenska handelshögskolan, 2003)
    Working Papers
    This paper uses panel unit root and cointegration methods to test the stationarity of the premium on domestic investors’ A shares over foreign investors’ B shares and cointegration between the A and B share prices on the Chinese stock exchanges. We find that the A share price premium is nonstationary until 2001, when the A and B share markets were partially merged, and that the A and B share prices are cointegrated in the panel.Cointegration is more likely to be found for firms in the service sector and for firms that issued B shares recently.