The document describes using support vector machines (SVMs) to improve assessment of deep vein thrombosis (DVT) compared to the standard Wells Score method. An SVM model was trained on 159 anonymous patient records containing Wells Score information and DVT findings. The SVM with an RBF kernel achieved 100% sensitivity and 58% accuracy, outperforming the Wells Score method. This demonstrates that machine learning can enhance clinical decision-making for DVT.
Tehnolabor is an engineering company with 15 years of experience developing electronic systems and technologies. They have 10 employees, 6 of whom are engineers, and had revenues of 1 million euros last year. Their clients come from various industries including museums, universities, and startups. Tehnolabor's goal is to realize their clients' ideas by developing innovative solutions using design, technology, and new approaches to problem solving. They offer services in hardware development, production management, and testing software. Notable past projects include interactive exhibits for museums, robotic kiosks, and remote control systems using mobile phones.
This document discusses the history and development of VideoLAN and the VLC media player. It began in 1995 as a student project to create a streaming media solution. In 1998 it was rebooted as an open source project with a modular, cross-platform design. Key developments included support for DVDs and Blu-Rays, network streaming, and ports to Android, iOS, and other platforms. Today VLC has over 1.2 billion downloads and contributions from over 600 developers. It remains fully open source and continues expanding to new formats and devices while maintaining a small, modular core.
This document discusses how the website SlideShare can help teachers provide educational presentations to students in a more engaging way than textbooks. It describes a specific presentation on writing literary analysis that defines important concepts and provides tables to help students organize their thoughts. The document notes that SlideShare allows students access presentations anywhere, helping to improve their work, and is more fun than textbooks. It also explains that while SlideShare hosts existing uploads, students could share presentations they create.
Se muestran algunas métodos de aprendizaje que se pueden utilizar en las fases predinámicas, dinámicas y posdinámicas; para la materia Evolución de la agricultura.
BEST is a non-profit organization that provides educational opportunities for engineering students across Europe. It organizes short-term courses on technical topics to help students develop skills in their field of study or for their future careers. This year, BEST will organize an advanced winter course in mobile development in Kosice, Slovakia for 20 foreign and 15 local students. The 8-day course will teach the basics of Android development and have students work in teams to build mobile apps.
Tehnolabor is an engineering company with 15 years of experience developing electronic systems and technologies. They have 10 employees, 6 of whom are engineers, and had revenues of 1 million euros last year. Their clients come from various industries including museums, universities, and startups. Tehnolabor's goal is to realize their clients' ideas by developing innovative solutions using design, technology, and new approaches to problem solving. They offer services in hardware development, production management, and testing software. Notable past projects include interactive exhibits for museums, robotic kiosks, and remote control systems using mobile phones.
This document discusses the history and development of VideoLAN and the VLC media player. It began in 1995 as a student project to create a streaming media solution. In 1998 it was rebooted as an open source project with a modular, cross-platform design. Key developments included support for DVDs and Blu-Rays, network streaming, and ports to Android, iOS, and other platforms. Today VLC has over 1.2 billion downloads and contributions from over 600 developers. It remains fully open source and continues expanding to new formats and devices while maintaining a small, modular core.
This document discusses how the website SlideShare can help teachers provide educational presentations to students in a more engaging way than textbooks. It describes a specific presentation on writing literary analysis that defines important concepts and provides tables to help students organize their thoughts. The document notes that SlideShare allows students access presentations anywhere, helping to improve their work, and is more fun than textbooks. It also explains that while SlideShare hosts existing uploads, students could share presentations they create.
Se muestran algunas métodos de aprendizaje que se pueden utilizar en las fases predinámicas, dinámicas y posdinámicas; para la materia Evolución de la agricultura.
BEST is a non-profit organization that provides educational opportunities for engineering students across Europe. It organizes short-term courses on technical topics to help students develop skills in their field of study or for their future careers. This year, BEST will organize an advanced winter course in mobile development in Kosice, Slovakia for 20 foreign and 15 local students. The 8-day course will teach the basics of Android development and have students work in teams to build mobile apps.
Social media: cosa dovrebbero sapere le aziende italianeNicoletta Boldrini
Presentazione tenuta in apertura del corso di Social Media Marketing #SMM360 - promosso da Roberto Gerosa, SocialDaily.it, e Nadia Zabbeo, Digysit.com - per professionisti e PMI. Dopo aver presentato alcuni dati utili a comprendere il panorama dei social oggi, ho voluto portare all'attenzione del pubblico alcune considerazioni su ciò che sta avvenendo a livello globale nelle modalità di comunicazione ed interazione delle persone... fenomeni che a mio avviso le aziende non possono ignorare.
Presentation in Japanese at Japan-Swiss symposium, 11 Oct 2014, at Keio University Hiyoshi campus. Link to the event: http://lib-arts.hc.keio.ac.jp/event/493
SlideShare permite compartir presentaciones de forma visual y esquemática para apoyar la educación. Las presentaciones subidas a SlideShare pueden ser utilizadas por docentes y estudiantes para explicar temas en clase o ser rescatadas más adelante por cualquier persona interesada. El registro y carga de presentaciones en SlideShare es un proceso sencillo que permite compartir fácilmente presentaciones con otros de manera más efectiva que enviar archivos por correo.
Rassegna della normativa sui tirocini in SiciliaGiovanni Nocera
In seguito alla pubblicazione delle linee guida garanzia giovani sicilia si ripropongono le slides di approfondimento sulla normativa regionale per l'attivazione dei tirocini formativi.
Este documento trata sobre la ética jurídica y sus principales conceptos. Explora las definiciones de moral, ética y deontología jurídica, así como las corrientes éticas más importantes como la de Sócrates, Platón, Aristóteles y el cristianismo. También cubre la ética personal y social, incluyendo valores, virtudes y libertad. Por último, analiza la ética en el ámbito administrativo, legislativo y judicial, así como la ética profesional en el mundo contemporáneo.
Corporate sponsorship involves costs associated with a project or program in exchange for recognition through logos and brand names displayed alongside the organization. For sponsorship, there must be relevance and brand fit between the sponsoring company and event, no conflict with the company's mission, and it should increase company or brand awareness to foster a positive view of the sponsoring company, even if no direct profit is produced.
The document summarizes the Harry Potter book series by J.K. Rowling. It provides biographical details about Rowling and describes the 7 Harry Potter books. Key characters like Harry, Hermione, Ron, Draco, Snape, Dumbledore, and Voldemort are outlined. The world of Hogwarts, houses, Quidditch, and more are briefly introduced. Over 300 million copies of the books have been sold worldwide after being translated into over 60 languages.
Blis insights: the influence of location on cross-screen advertisingBlis
Understanding the role of location in how consumers interact and respond to advertising is becoming a critical component for every brand looking to execute successful mobile-aware campaigns.
Technology has played its part in this evolution, with devices and connectivity combining to provide consumers with access to content and information anytime, anyplace. Consumers are using multiple devices to interact with brands in different ways depending on the kind of content being consumed, the services they are using, and the context of their location.
The data in this deck comes from a Blis White Paper based on consumer research from a nationally represented sample of UK consumers, designed to explore the following topics:
Device ownership and usage across demographics
How consumers engage with mobile advertising on different devices
The importance of context in mobile advertising
The influence of location on mobile advertising engagement
This document proposes using machine learning techniques to predict COVID-19 infections based on chest x-ray images. Specifically, it involves using discrete wavelet transform to extract space-frequency features from chest x-rays, reducing the dimensionality of features using Shannon entropy, and then training standard machine learning classifiers like logistic regression, support vector machine, decision tree, and convolutional neural network on the extracted features to classify images as COVID-19 positive or negative. The document provides background on the proposed techniques of discrete wavelet transform, entropy, and various machine learning models.
This document provides an introduction to multivariate data analysis (MVA) using R. It defines what multivariate analysis is and explains that multivariate datasets contain multiple variables and can include mixed data types. Common MVA methods are discussed, including hierarchical cluster analysis (HCA) and partition clustering for exploratory analysis. HCA involves calculating distances between variables, building a tree to visualize relationships, and can identify potential subgroups. Partition clustering assigns variables to discrete clusters. The document demonstrates HCA and partition clustering on gene expression data to explore patterns among patients and genes.
Healthcare deserts: How accessible is US healthcare?Data Con LA
Data Con LA 2020
Description
In 2018, healthcare spending in the US accounted for 17% of the nation’s GDP. With such significant spend, how can we better understand what that means for healthcare and treatment accessibility? When policy changes occur, how can we gauge the impact on rural areas, which are disproportionally affected by inadequate access to healthcare (or “healthcare deserts”)? Using publicly available data and records, it is possible to locate all major hospitals in the U.S. and, for every residential ZIP code, model the population affected by healthcare deserts at various travel mileage thresholds. This talk will focus on:
· The several public datasets that are available to address this question
· The logic and algorithm(s) used to compute this efficiently in Python
· Visualizing the problem and telling the story in Tableau
Speaker
Andrew Kaszpurenko, Edwards Lifesciences, Manager of Advanced Analytics at Edwards Lifesciences THV Division
IRJET- Diagnosis of Diabetic Retinopathy using Machine Learning AlgorithmsIRJET Journal
This document discusses using machine learning algorithms to diagnose diabetic retinopathy through analysis of retinal images. Specifically, it proposes using a support vector machine (SVM) to classify images based on features extracted from image processing techniques. An initial image processing stage isolates features like blood vessels and lesions. The SVM then analyzes these features to determine the grade of diabetic retinopathy. Stochastic gradient descent is also used, achieving a classification accuracy of 91%. The goal is to automatically detect diabetic retinopathy at early stages to prevent vision loss.
1) The document describes the development, verification, and validation of a responsive boundary model.
2) Verification was performed using manufactured solutions and grid convergence studies to evaluate spatial and temporal discretization errors.
3) Validation compared model predictions of helium plume interferometry data to experimental measurements, analyzing sources of error and sensitivity.
4) Results showed good agreement between computation and experiment, demonstrating the model captured important physical phenomena.
A Multi-Objective Genetic Algorithm for Pruning Support Vector MachinesMohamed Farouk
This document summarizes research on using a multi-objective genetic algorithm to prune support vectors from support vector machines. Experiments on four datasets showed the approach could reduce computational complexity by 63-78% by reducing the number of support vectors, without sacrificing training accuracy and sometimes improving test set accuracy. Future work plans to extend the approach to support vector regression and test additional kernel functions.
Social media: cosa dovrebbero sapere le aziende italianeNicoletta Boldrini
Presentazione tenuta in apertura del corso di Social Media Marketing #SMM360 - promosso da Roberto Gerosa, SocialDaily.it, e Nadia Zabbeo, Digysit.com - per professionisti e PMI. Dopo aver presentato alcuni dati utili a comprendere il panorama dei social oggi, ho voluto portare all'attenzione del pubblico alcune considerazioni su ciò che sta avvenendo a livello globale nelle modalità di comunicazione ed interazione delle persone... fenomeni che a mio avviso le aziende non possono ignorare.
Presentation in Japanese at Japan-Swiss symposium, 11 Oct 2014, at Keio University Hiyoshi campus. Link to the event: http://lib-arts.hc.keio.ac.jp/event/493
SlideShare permite compartir presentaciones de forma visual y esquemática para apoyar la educación. Las presentaciones subidas a SlideShare pueden ser utilizadas por docentes y estudiantes para explicar temas en clase o ser rescatadas más adelante por cualquier persona interesada. El registro y carga de presentaciones en SlideShare es un proceso sencillo que permite compartir fácilmente presentaciones con otros de manera más efectiva que enviar archivos por correo.
Rassegna della normativa sui tirocini in SiciliaGiovanni Nocera
In seguito alla pubblicazione delle linee guida garanzia giovani sicilia si ripropongono le slides di approfondimento sulla normativa regionale per l'attivazione dei tirocini formativi.
Este documento trata sobre la ética jurídica y sus principales conceptos. Explora las definiciones de moral, ética y deontología jurídica, así como las corrientes éticas más importantes como la de Sócrates, Platón, Aristóteles y el cristianismo. También cubre la ética personal y social, incluyendo valores, virtudes y libertad. Por último, analiza la ética en el ámbito administrativo, legislativo y judicial, así como la ética profesional en el mundo contemporáneo.
Corporate sponsorship involves costs associated with a project or program in exchange for recognition through logos and brand names displayed alongside the organization. For sponsorship, there must be relevance and brand fit between the sponsoring company and event, no conflict with the company's mission, and it should increase company or brand awareness to foster a positive view of the sponsoring company, even if no direct profit is produced.
The document summarizes the Harry Potter book series by J.K. Rowling. It provides biographical details about Rowling and describes the 7 Harry Potter books. Key characters like Harry, Hermione, Ron, Draco, Snape, Dumbledore, and Voldemort are outlined. The world of Hogwarts, houses, Quidditch, and more are briefly introduced. Over 300 million copies of the books have been sold worldwide after being translated into over 60 languages.
Blis insights: the influence of location on cross-screen advertisingBlis
Understanding the role of location in how consumers interact and respond to advertising is becoming a critical component for every brand looking to execute successful mobile-aware campaigns.
Technology has played its part in this evolution, with devices and connectivity combining to provide consumers with access to content and information anytime, anyplace. Consumers are using multiple devices to interact with brands in different ways depending on the kind of content being consumed, the services they are using, and the context of their location.
The data in this deck comes from a Blis White Paper based on consumer research from a nationally represented sample of UK consumers, designed to explore the following topics:
Device ownership and usage across demographics
How consumers engage with mobile advertising on different devices
The importance of context in mobile advertising
The influence of location on mobile advertising engagement
This document proposes using machine learning techniques to predict COVID-19 infections based on chest x-ray images. Specifically, it involves using discrete wavelet transform to extract space-frequency features from chest x-rays, reducing the dimensionality of features using Shannon entropy, and then training standard machine learning classifiers like logistic regression, support vector machine, decision tree, and convolutional neural network on the extracted features to classify images as COVID-19 positive or negative. The document provides background on the proposed techniques of discrete wavelet transform, entropy, and various machine learning models.
This document provides an introduction to multivariate data analysis (MVA) using R. It defines what multivariate analysis is and explains that multivariate datasets contain multiple variables and can include mixed data types. Common MVA methods are discussed, including hierarchical cluster analysis (HCA) and partition clustering for exploratory analysis. HCA involves calculating distances between variables, building a tree to visualize relationships, and can identify potential subgroups. Partition clustering assigns variables to discrete clusters. The document demonstrates HCA and partition clustering on gene expression data to explore patterns among patients and genes.
Healthcare deserts: How accessible is US healthcare?Data Con LA
Data Con LA 2020
Description
In 2018, healthcare spending in the US accounted for 17% of the nation’s GDP. With such significant spend, how can we better understand what that means for healthcare and treatment accessibility? When policy changes occur, how can we gauge the impact on rural areas, which are disproportionally affected by inadequate access to healthcare (or “healthcare deserts”)? Using publicly available data and records, it is possible to locate all major hospitals in the U.S. and, for every residential ZIP code, model the population affected by healthcare deserts at various travel mileage thresholds. This talk will focus on:
· The several public datasets that are available to address this question
· The logic and algorithm(s) used to compute this efficiently in Python
· Visualizing the problem and telling the story in Tableau
Speaker
Andrew Kaszpurenko, Edwards Lifesciences, Manager of Advanced Analytics at Edwards Lifesciences THV Division
IRJET- Diagnosis of Diabetic Retinopathy using Machine Learning AlgorithmsIRJET Journal
This document discusses using machine learning algorithms to diagnose diabetic retinopathy through analysis of retinal images. Specifically, it proposes using a support vector machine (SVM) to classify images based on features extracted from image processing techniques. An initial image processing stage isolates features like blood vessels and lesions. The SVM then analyzes these features to determine the grade of diabetic retinopathy. Stochastic gradient descent is also used, achieving a classification accuracy of 91%. The goal is to automatically detect diabetic retinopathy at early stages to prevent vision loss.
1) The document describes the development, verification, and validation of a responsive boundary model.
2) Verification was performed using manufactured solutions and grid convergence studies to evaluate spatial and temporal discretization errors.
3) Validation compared model predictions of helium plume interferometry data to experimental measurements, analyzing sources of error and sensitivity.
4) Results showed good agreement between computation and experiment, demonstrating the model captured important physical phenomena.
A Multi-Objective Genetic Algorithm for Pruning Support Vector MachinesMohamed Farouk
This document summarizes research on using a multi-objective genetic algorithm to prune support vectors from support vector machines. Experiments on four datasets showed the approach could reduce computational complexity by 63-78% by reducing the number of support vectors, without sacrificing training accuracy and sometimes improving test set accuracy. Future work plans to extend the approach to support vector regression and test additional kernel functions.
bayesImageS: Bayesian computation for medical Image Segmentation using a hidd...Matt Moores
This document summarizes an R package called bayesImageS that enables Bayesian computation for medical image segmentation using a hidden Potts model. It discusses the statistical model, which involves a hidden Markov random field with a Potts prior on the latent labels. Bayesian computation methods like Gibbs sampling and Metropolis-Hastings using pseudolikelihood approximation are implemented in C++ for efficiency. Experimental results demonstrate the package on a CT electron density phantom and patient radiotherapy data.
Abdominal Bleeding Region Detection in Wireless Capsule Endoscopy Videos Usin...journal ijrtem
Engineering and Technology, Kandala
Abstract: Wireless Capsule Endoscopy (WCC) is a medical imaging technique used to examine parts of the gastrointestinal tract.
Computer aided detection is used to increase the speed of detection, better performance and reduce the time. Before finding the
bleeding regions the edge regions are first detected and removed. Both the edge and the bleeding regions will share the same Hue
value and the luminance should be same for the bleeding and the non -bleeding regions .We use a canny edge detector operator for
detecting the edge regions in L channel. Canny edge detector is used to detect more edge pixels and preserve more bleeding pixels
based up on canny edge algorithm. This method in edge removal algorithm includes edge detection, edge dilation and edge masking.
After the removal of edges, those regions are made in to segment through super-pixel segmentation and regions are classified using
Artificial Neural Network by Radial Bias Function (RBF).
Keywords: Canny edge detector, edge detection, super-pixel segmentation.
Breast cancer diagnosis and recurrence prediction using machine learning tech...eSAT Journals
Abstract Breast Cancer has become the common cause of death among women. Due to long hours invested in manual diagnosis and lesser diagnostic system available emphasize the development of automated diagnosis for early diagnosis of the disease. Our aim is to classify whether the breast cancer is benign or malignant and predict the recurrence and non-recurrence of malignant cases after a certain period. To achieve this we have used machine learning techniques such as Support Vector Machine, Logistic Regression, KNN and Naive Bayes. These techniques are coded in MATLAB using UCI machine learning depository. We have compared the accuracies of different techniques and observed the results. We found SVM most suited for predictive analysis and KNN performed best for our overall methodology. Keywords: Breast Cancer, SVM, KNN, Naive Bayes, Logistic Regression, Classification.
Encoder for (7,3) cyclic code using matlabSneheshDutta
This document provides an overview of cyclic codes including:
- What cyclic codes are and their properties of error detection and correction.
- The method of generating cyclic codes by multiplying message polynomials by a generator polynomial.
- How to systematically encode cyclic codes in three steps.
- The encoding and decoding circuits including Meggitt decoder.
- An example of a (7,3) cyclic code implemented in Matlab showing the encoding, corruption with errors, and decoding.
- How cyclic codes can detect errors through syndrome computation and lookup tables.
- Applications of cyclic codes in message identification.
Improving patient care along with quality service and at the same time cutting costs might seem totally opposite goals, but that is exactly the promise of telemedicine. The term telemedicine covers an ever increasing number of techniques that put patients in touch with medical expertise over great distances. Techlead Software Engineering Pvt. Ltd. India is coming up with one stop solution for ophthalmologist, the Digitech Ophthalmologist.
Digitech Ophthalmologist
Digital ophthalmologist is a software tool developed by Techlead software Engineering. It has been developed with understanding of the ophthalmologist’s concerns. This smart assistant to doctors will perform many of their routine and time consuming tasks in a much efficient way.
Panorama ConstructionCDR AnalysisGlaucoma DetectionBlood Vessel Analysis3D Retina ReconstructionRetinopathy
This document discusses statistical analysis of measurement data. It provides examples of calculating the mean, median, variance, and standard deviation of data sets. Calculating these statistical values allows determination of the uncertainty in measurement results and confidence in the mean or median values. As the number of measurements increases and variance and standard deviation decrease, confidence in the mean value being close to the true value also increases. Microsoft Excel functions to calculate sum, average, average deviation, standard deviation are also presented.
Teleradiology Overview Systems and Applications - Sanjoy SanyalSanjoy Sanyal
Teleradiology allows radiographic images and reports to be electronically transmitted between locations. It consists of an image sending station, transmission network, and receiving/image review station. Common applications include allowing off-site radiologists and specialists to review images and providing consultations for rural primary care physicians. Key factors in teleradiology systems include image resolution, compression methods, and sufficient transmission speeds over networks like point-to-point, local area networks, or wide area networks.
Cloud computing can be a useful tool to analyze bio signals; when that methodology is combined with the over-the-air transfer of data, it allows healthcare providers to access relevant information on a lightweight mobile interface. Here, we develop an Android application that receives an ECG signal over Bluetooth, plots the data stream, and allows a user to send a biosignal analysis request to determine if any abnormality is present in the ECG signal. The application can receive the result of the analysis within seconds of time, allowing healthcare providers to efficiently analyze incoming data.
This document discusses using unsupervised support vector analysis to increase the efficiency of simulation-based functional verification. It describes applying an unsupervised machine learning technique called support vector analysis to filter redundant tests from a set of verification tests. By clustering similar tests into regions of a similarity metric space, it aims to select the most important tests to verify a design while removing redundant tests, improving verification efficiency. The approach trains an unsupervised support vector model on an initial set of simulated tests and uses it to filter future tests by comparing them to support vectors that define regions in the similarity space.
(short presentation) A Wearable Monitoring System for Continuously Assessing ...Meghan Hegarty
This document summarizes research on developing a wearable monitoring system to continuously assess peripheral vascular health. It describes work on measuring interface pressure using sensor arrays, blood flow velocity using Doppler ultrasound, and leg volume using impedance plethysmography. The goal is to create minimally invasive, wireless devices that can provide near real-time measurements to improve chronic venous disease diagnosis and treatment monitoring.
A Wearable Monitoring System for Continuously Assessing the Health of the Per...Meghan Hegarty
MedTech 2010
M. Hegarty, F. Livingston, E. Grant, L. Reid
Center for Robotics and Intelligent Machines (North Carolina State University)
Carolon Company (Rural Hall, NC)
Interpretation of electrocardiography (ECG) by using polynomial function simu...mustafatacettin1
This document describes a study that uses polynomial function simulation and random forest algorithms to interpret electrocardiography (ECG) data and predict heart diseases. The study aims to develop a more accurate machine learning model for detecting rhythm-related heart diseases from ECG data. The model calculates unique parameters from the ECG data, including average QRS length, polynomial differences, entropy values, and P wave direction. When tested on a dataset of over 10,000 ECGs, the model achieves high prediction accuracy for various heart diseases such as atrial fibrillation, sinus bradycardia, and supraventricular tachycardia. A software interface is also developed to allow users to input ECG data and receive disease
The document presents a new method called KCGex-SVM for extracting rules from support vector machines (SVMs). It combines weighted kernel k-means clustering, genetic algorithms, and information from SVMs to generate an interpretable rule set from credit screening data. The method was tested on three credit screening datasets and showed improved accuracy over other rule extraction techniques, generating rules with good performance while maintaining comprehensibility.
3. Why?
Every year 1.6 per 1000 inhabitants suffer from venous
thrombosis, blood cloths in their veins.
Commonly assessed by a combination of D-Dimer tests and a
questionnaire.
Can we make a system that makes the assessment better?
4. D-Dimer
A small protein fragment present in the blood after a blood
clot is degraded. High concentration of D-Dimer correlates
with thrombosis, but false positive readings can occur from
liver disease, high rheumatoid factor, inflammation, and many
other factors which makes the sensitivity low.
5. Wells Score
Gold standard for DVT clinical assessment, named Wells Score.
A simple scoring system and a yes or no questionnaire
regarding the patients medical history.
Very quick assessment as it only has 9 significant variables.
6. Assessment in real world
A combination of Wells Score and a D-dimer test reliably
excludes DVT
Often no need for painful imaging studies such as compression
ultrasonography
7. Compression Ultrasonography
Compressing the veins
Send pulses of ultrasound into the leg to find material with
different density
Note absence or presence of occluded veins
Needed to be reported
9. Deep Vein Thrombosis
Blood is constantly coagulating and dissolving.
This balance between the stimulating and inhibitory is sensitive.
Coagulated blood could produce blood clots
Dangerous internal bleeding otherwise
10. Rate of occurrence
If part of a cloth breaks free and is allowed to travel to the lung , we
call it a pulmonary embolism . A serious condition that 1000
patients a year die from just in Sweden. These figures should be
compared to the ~4000 who gets diagnosed with pulmonary
embolism or ~8000 patients who gets diagnosed with deep vein
thrombosis.
11. Costs
The Swedish hospitals costs of venous thromboembolism alone
was estimated to 0.375 billion SEK in 1999.
Difficulty to confirm diagnosis without thrombosonography
12. Pretest
The decision to order thrombosonography is, in several
guidelines, in large part done based entirely on pretest risk
assessment like Wells Score . The patients with low risk get
D-dimer blood test and only go on to ultrasonography if the
test is positive. The ones with high risk goes straight to
ultrasonography without getting a D-dimer test.
14. Can we improve Wells Score?
With machine learning algorithms we can figure out how to
assess deep vein thrombosis by generalising from examples.
As more data would become available in the reporting system
the better the assessment would become.
15. Support Vector Machines
Developed by Cortes & Vapnik for classification.
Finding the maximal margin of separation between two classes,
which can be seen as the generalisation, for linearly separable
patterns.
16. Support Vector Machines
With support vector classifiers the hyperplane can be described by
the unknown, u, and the margin-vector, w (see figure 2):
w · u ≥ C (1)
18. Support Vector Machines
We are interested in which side the u vector is in so we project the
u onto w and if this is bigger than some constant C then we say
that u a positive sample.
19. Support Vector Machines
By setting b = −C we get our decision rule:
(w · u) + b = 0, w ∈ RN
, u ∈ RN
, b ∈ R (2)
which corresponds to the decision function:
f (x) = sign((w · u) + b) (3)
20. Support Vector Machines
The problem is that we do not know neither the w nor the b.
However adding additional constraints we can calculate them.
Taking a positive and negative sample and arbitrary setting it to
bigger and smaller than one respectively:
w · x+ ≥ 1 (4)
w · x− ≤ 1 (5)
21. Support Vector Machines
Then introducing a variable yi that is +1 for positive samples and
−1 for negative samples we can combine these into:
yi ∗ (xi · w + b) ≥ 1 (6)
22. Support Vector Machines
And with that we can add the extra constraint that:
yi ∗ (xi · w + b) − 1 = 0 (7)
should be were xi is on the margin.
23. Support Vector Machines
Now if we want the widest margin possible we could take the
difference of a negative and positive sample on the margins and
project it onto a unit normal.
WIDTH = (x+ − x−) ·
w
w
(8)
25. Support Vector Machines
The way one maximize this in the support vector machines
algorithm is to use Lagrange multipliers.
26. Radial Basis Function
But with the so called kernel-trick , were we map data into a richer
feature space then construct a hyperplane in that space, we are able
to classify points that were not linearly separable in its previous
space.
We call the function that maps from the vector x to another input
space φ(x).
27. Radial Basis Function
By doing this simple transformation we know need to maximize:
K(x, y) = φ(x) · φ(y) (10)
28. Radial Basis Function
Don’t need φ(x) on its own but can instead focus on K(x, y) which
we call our kernel function.
By using the radial basis kernel (RBF):
K(x, y) = e−γ x−y 2
(11)
where γ is a chosen constant, we get a great and fast nonlinear
kernel.
29. Environment
Apple iOS
C++ and Objective-C
OpenCV developed by Intel Russia research center in Nizhny
Novgorod for realtime computer vision. This library contains
implementations for both RBF and linear kernels as it adopted
the SVM/C++ library libsvm.
30. Preprocessing
159 anonymous patients DVT journals
From the journals we extracted the Wells score information and
whether a DVT or occlusion were found.
Table 1:Count of each label in dataset
DVT Nothing found
33 126
Already gone through a Wells score, heavy bias.
Yes and no converted to 1.0 and -1.0 respectively.
31. Training
Optimized C and gamma values
5 folds
C-values between 2−5 and 215
Gamma-values between 2−15 and 23
33. Baseline
Table 2:Wells score
Variable
Points
Cancer treatment during the past 6 months +1
Lower leg paralysis or plastering +1
Bed rest > 3 days or surgery < 4 weeks +1
Pain on palpation of deep veins +1
Swelling of entire leg +1
Diameter difference on affected calf > 3 cm +1
Pitting oedema (affected side only) +1
Dilated superficial veins (affected side) +1
Alternative diagnosis at least as probable as DVT -2
45. Results
Class Weight Accuracy
SVM RBF 0.9226 58.49%
Wells Score - MEDIUM N/A 23.12%
SVM RBF 0.8196 81.11%
SVM Linear 0.9193 65.40%
Wells Score - HIGH N/A 58.49%
46. Results
Class Weight Sensitivity Specificity
SVM RBF 0.9226 100.00% 35.71%
Wells Score - MEDIUM N/A 97.05% 3.17%
SVM RBF 0.8196 66.66% 84.92%
SVM Linear 0.9193 63.63% 65.87%
Wells Score - HIGH N/A 60.60% 57.93%
47. Balanced Error Rate
BER =
FP/(TN + FP) + FN/(FN + TP)
2
Average of both the error rate of the positive class and the error
rate of the negative class.
48. Diagnostic Odds Ratio
DOR =
TP/FP
FN/TN
Ratio of the odds of the test being positive if the subject has a
disease relative to the odds of the test being positive if the subject
does not have the disease.
49. Results
Class Weight BCR DOR
SVM RBF 0.9226 67.85% ∞
Wells Score - MEDIUM N/A 50.11% 01.08
SVM RBF 0.8196 75.79% 11.26
SVM Linear 0.9193 64.75% 03.37
Wells Score - HIGH N/A 59.27% 02.11
50. Conclusion
Possible to improve DVT assessment by using SVMs.
Able to get 100% sensitivity with 58% accuracy.
Balanced Classification Rate for Well Score was at most 02.11
whilst our highest benchmarked was 11.26 .
Using different error costs (DEC) we can tweak the sensitivity
and specificity.