Classification of cancer-related death certificates using machine learning

Background Cancer monitoring and prevention relies on the critical aspect of timely notification of cancer cases. However, the abstraction and classification of cancer from the free-text of pathology reports and other relevant documents, such as death certificates, exist as complex and time-consuming activities. Aims In this paper, approaches for the automatic detection of notifiable cancer cases as the cause of death from free-text death certificates supplied to Cancer Registries are investigated. Method A number of machine learning classifiers were studied. Features were extracted using natural language techniques and the Medtex toolkit. The numerous features encompassed stemmed words, bi-grams, and concepts from the SNOMED CT medical terminology. The baseline consisted of a keyword spotter using keywords extracted from the long description of ICD-10 cancer related codes. Results Death certificates with notifiable cancer listed as the cause of death can be effectively identified with the methods studied in this paper. A Support Vector Machine (SVM) classifier achieved best performance with an overall F-measure of 0.9866 when evaluated on a set of 5,000 free-text death certificates using the token stem feature set. The SNOMED CT concept plus token stem feature set reached the lowest variance (0.0032) and false negative rate (0.0297) while achieving an F-measure of 0.9864. The SVM classifier accounts for the first 18 of the top 40 evaluated runs, and entails the most robust classifier with a variance of 0.001141, half the variance of the other classifiers. Conclusion The selection of features significantly produced the most influences on the performance of the classifiers, although the type of classifier employed also affects performance. In contrast, the feature weighting schema created a negligible effect on performance. Specifically, it is found that stemmed tokens with or without SNOMED CT concepts create the most effective feature when combined with an SVM classifier.

An exploration into retrofitting cities with natural design elements - learning from the city of Berlin, Germany

Natural design features in the built environment or biophilic elements are emerging as a potential response to the challenges of climate change, urbanisation and population pressures which have invited issues such as rising urban heat island effect, rising pollution, increased congestion, among others. This concept of living cities was made popular by Professor Tim Beatley in his book titled ‘Biophilic Urbanism’. Evidence of biophilic urbanism can be seen in some cities from around the globe since decoupling environmental pressures from future development is a priority on many agendas. Berlin is an example of a modern economy that has adopted an ecological sustainable development approach to reduce environmental degradation while driving innovation and employment.

An alternate learning approach for destructive testing in civil engineering - Benefits from remote laboratory experimentation

An alternative learning approach for destructive testing of structural specimens in civil engineering is explored by using a remote laboratory experimentation method. The remote laboratory approach focuses on overcoming the constraints in the hands-on experimentation without compromising the understanding of the students on the concepts and mechanics of reinforced concrete structures. The goal of this study is to evaluate whether or not the remote laboratory experimentation approach can become a standard in civil engineering teaching. The teaching activity using remote-laboratory experimentation is presented here and the outcomes of this activity are outlined. The experience and feedback gathered from this study are used to improve the remote-laboratory experimentation approach in future years to other aspects of civil engineering where destructive testing is essential.

Developing learning cohorts for postgraduate research degrees

Development of researchers through higher degree research studies is a high priority in most universities. Yet, research about supervision as pedagogy and models of supervision is only recently gained increasing attention. Charged with producing good researchers within very limited resources, academics are constantly looking for more efficient models of supervision for higher degree research students. A cohort model of supervision promises several efficiencies, but we argue that its success lies importantly on how well the cohort is developed specifically for higher degree research studies. We drew on a growing body of literature on higher degree research supervision to design, implement and evaluate our approach to developing a cohort of seven students enrolled in the Master of Education (Research) degree. Our approach included four provisions: initial residential workshop, development of a learning community, nourishing scholarship, and ongoing learning opportunities. The four provisions resulted in gradually developing an environment and culture that students found very supportive and nurturing. This paper is based on the findings from data collected from student evaluations in the first year of studies, feedback from the cohort’s sponsor, and our reflective notes. The evaluation substantiated the value in investing time and resources for purposely developing a cohort for higher degree research studies. Whether the cohorts are sponsored or not, universities will still need to invest time and resources for cohort development if a cohort model is intended to gain wider efficiencies in supervision of higher degree research students.

Applying threshold learning theory to teach sustainable business practice in post-graduate engineering education

This paper presents the results of a qualitative action-research inquiry into how a highly diverse cohort of post-graduate students could develop significant capacity in sustainable development within a single unit (course), in this case a compulsory component of four built environment masters programs. The method comprised applying threshold learning theory within the technical discipline of sustainable development, to transform student understanding of sustainable business practice in the built environment. This involved identifying a number of key threshold concepts, which once learned would provide a pathway to having a transformational learning experience. Curriculum was then revised, to focus on stepping through these targeted concepts using a scaffolded, problem-based-learning approach. Challenges included a large class size of 120 students, a majority of international students, and a wide span of disciplinary backgrounds across the spectrum of built environment professionals. Five ‘key’ threshold learning concepts were identified and the renewed curriculum was piloted in Semester 2 of 2011. The paper presents details of the study and findings from a mixed-method evaluation approach through the semester. The outcomes of this study will be used to inform further review of the course in 2012, including further consideration of the threshold concepts. In future, it is anticipated that this case study will inform a framework for rapidly embedding sustainability within curriculum.

Enhancing learning in large undergraduate classes using custom made audio-visual cases

This is an exploratory study into the effective use of embedding custom made audiovisual case studies (AVCS) in enhancing the student’s learning experience. This paper describes a project that used AVCS for a large divergent cohort of undergraduate students, enrolled in an International Business course. The study makes a number of key contributions to advancing learning and teaching within the discipline. AVCS provide first hand reporting of the case material, where the students have the ability to improve their understanding from both verbal and nonverbal cues. The paper demonstrates how AVCS can be embedded in a student-centred teaching approach to capture the students’ interest and to enhance a deep approach to learning by providing real-world authentic experience.

Application of machine learning technique in wind turbine fault diagnosis

In this study, a machine learning technique called anomaly detection is employed for wind turbine bearing fault detection. Basically, the anomaly detection algorithm is used to recognize the presence of unusual and potentially faulty data in a dataset, which contains two phases: a training phase and a testing phase. Two bearing datasets were used to validate the proposed technique, fault-seeded bearing from a test rig located at Case Western Reserve University to validate the accuracy of the anomaly detection method, and a test to failure data of bearings from the NSF I/UCR Center for Intelligent Maintenance Systems (IMS). The latter data set was used to compare anomaly detection with SVM, a previously well-known applied method, in rapidly finding the incipient faults.

Using visualization tool to help engineering students learning dynamics

Dynamics is an essential core engineering subject. It includes high level mathematical and theoretical contents, and basic concepts which are abstract in nature. Hence, Dynamics is considered as one of the hardest subjects in the engineering discipline. To assist our students in learning this subject, we have conducted a Teaching & Learning project to study ways and methods to effectively teach Dynamics based on visualization techniques. The research project adopts the five basic steps of Action Learning Cycle. It is found that visualization technique is a powerful tool for students learning Dynamics and helps to break the barrier of students who perceived Dynamics as a hard subject.