I have a Doctorate in Biotechnology and I'm a machine learning expert based in Hong Kong.
Novel AI study suggests we can know if we're still around in 5 years time. By scanning our organs.
A proper prediction of patient's longevity has been hampered by doctors' inability to measure the health of each organ. A recent study published in the Nature journal Scientific Reports. Where artificial intelligence was used to analyse the images of 48 patients' organs, with a 69% certainty which patients will die in five years. This is comparable to the accuracy of human doctors in predicting a patient's lifespan.
Researchers applied feature engineering and deep learning methods to predict the age of a person. Early diagnose of critical illness is valuable as it empowers doctors with medical insights for designing customised treatment to prolong life. medical interventions
The team performed a backward-looking case-control study, with matching used to control for non-imaging clinical and demographic variables that were expected to be highly predictive of five-year mortality rate. Using training data sets they assessed the predictive performance of the feature engineering and deep learning methodologies based on a 6-fold cross-validation experiment. A qualitative visual comparison was made between the cases correctly identified as mortality or survival cases by the deep learning and feature engineering models.
Another related development is the early detection of breast cancer. Survival rates of breast cancer characterised by early detection are 99% as opposed to only 27% with late diagnosis. The Intelligent Bra (iTBra) comprises of breast sensor patches embedded into bras, engineered to detect dynamic circadian temperature fluctuations of breast tissue.
In contrast with mammography and ultrasound, which rely on specular reflective capabilities of varying tissue densities, the solution is tissue agnostic and detects early circadian cellular changes in all tissue types and varied age groups. Dense breast tissue is comprised of less fat and more connective/fibrous and glandular tissue, and ranges in severity from Level A to Level D. As the density of breasts increases, the ability of the mammogram to detect cancer decreases.
The cancer risk in women with highly dense breast tissue can be up to 6 times higher compared to normal/fatty tissue and follows an acceleration of breast cancer. Using the iTBra means no more biopsies and avoids false positives.
Abstract: Personalized medicine is not a new concept. The renaissance of the term is due to the enormous progress in gene sequencing technology and functional imaging, as well as the development of targeted therapies. Application of these technologies in clinical medicine will necessitate infrastructural as well as organizational and educational changes in the healthcare system. An important change required already in the short-term is the introduction of centralized structures, preferably in university clinics, which adopt these innovations and incorporate them into clinical care. Simultaneously, the collation and use of large quantities of relevant data from highly variable sources must be successfully mastered, in order to pave the way for disruptive technologies such as artificial intelligence.
Pub.: 26 May '17, Pinned: 12 Jun '17
Abstract: Tissue biomarker scoring by pathologists is central to defining the appropriate therapy for patients with cancer. Yet, inter-pathologist variability in the interpretation of ambiguous cases can affect diagnostic accuracy. Modern artificial intelligence methods such as deep learning have the potential to supplement pathologist expertise to ensure constant diagnostic accuracy. We developed a computational approach based on deep learning that automatically scores HER2, a biomarker that defines patient eligibility for anti-HER2 targeted therapies in breast cancer. In a cohort of 71 breast tumour resection samples, automated scoring showed a concordance of 83% with a pathologist. The twelve discordant cases were then independently reviewed, leading to a modification of diagnosis from initial pathologist assessment for eight cases. Diagnostic discordance was found to be largely caused by perceptual differences in assessing HER2 expression due to high HER2 staining heterogeneity. This study provides evidence that deep learning aided diagnosis can facilitate clinical decision making in breast cancer by identifying cases at high risk of misdiagnosis.
Pub.: 06 Apr '17, Pinned: 09 Jun '17
Abstract: Precision medicine approaches rely on obtaining precise knowledge of the true state of health of an individual patient, which results from a combination of their genetic risks and environmental exposures. This approach is currently limited by the lack of effective and efficient non-invasive medical tests to define the full range of phenotypic variation associated with individual health. Such knowledge is critical for improved early intervention, for better treatment decisions, and for ameliorating the steadily worsening epidemic of chronic disease. We present proof-of-concept experiments to demonstrate how routinely acquired cross-sectional CT imaging may be used to predict patient longevity as a proxy for overall individual health and disease status using computer image analysis techniques. Despite the limitations of a modest dataset and the use of off-the-shelf machine learning methods, our results are comparable to previous 'manual' clinical methods for longevity prediction. This work demonstrates that radiomics techniques can be used to extract biomarkers relevant to one of the most widely used outcomes in epidemiological and clinical research - mortality, and that deep learning with convolutional neural networks can be usefully applied to radiomics research. Computer image analysis applied to routinely collected medical images offers substantial potential to enhance precision medicine initiatives.
Pub.: 12 May '17, Pinned: 09 Jun '17
Abstract: Publication date: Available online 20 August 2016 Source:Handbook of Statistics Author(s): F.D. Hudson, E.W. Nichols The Internet of Things (IoT) and Cognitive Computing are concepts that have been developing since the 1950s with varying nomenclature. Enabled by advances in technology such as the development of lower cost, lower power, smaller microprocessors and microcontrollers, cellular/wireless chips, and the enablement of more widespread machine-to-machine communications, IoT technologies are now more widely deployed. Advances in Artificial Intelligence (AI) and question-answering systems (Ferrucci et al., 2010; Simmons, 1970) have enabled development of cognitive computing systems which can be applied to numerous use cases. There is an opportunity to leverage IoT and cognitive computing technologies together with data analytics (http://social-innovation.hitachi.com) to develop deeper insights from the vast amount of data being created by IoT to create value for people, cities, and industry. In this chapter, we explore the state of the art and future opportunities to apply IoT and cognitive computing together in IoT uses cases for smart cities and connected healthcare, to improve efficiencies, outcomes, and the human experience.
Pub.: 25 Aug '16, Pinned: 12 Jun '17
Abstract: The International Symposium on Biomedical Imaging (ISBI) held a grand challenge to evaluate computational systems for the automated detection of metastatic breast cancer in whole slide images of sentinel lymph node biopsies. Our team won both competitions in the grand challenge, obtaining an area under the receiver operating curve (AUC) of 0.925 for the task of whole slide image classification and a score of 0.7051 for the tumor localization task. A pathologist independently reviewed the same images, obtaining a whole slide image classification AUC of 0.966 and a tumor localization score of 0.733. Combining our deep learning system's predictions with the human pathologist's diagnoses increased the pathologist's AUC to 0.995, representing an approximately 85 percent reduction in human error rate. These results demonstrate the power of using deep learning to produce significant improvements in the accuracy of pathological diagnoses.
Pub.: 18 Jun '16, Pinned: 09 Jun '17
Abstract: Recent advances in machine learning yielded new techniques to train deep neural networks, which resulted in highly successful applications in many pattern recognition tasks such as object detection and speech recognition. In this paper we provide a head-to-head comparison between a state-of-the art in mammography CAD system, relying on a manually designed feature set and a Convolutional Neural Network (CNN), aiming for a system that can ultimately read mammograms independently. Both systems are trained on a large data set of around 45,000 images and results show the CNN outperforms the traditional CAD system at low sensitivity and performs comparable at high sensitivity. We subsequently investigate to what extent features such as location and patient information and commonly used manual features can still complement the network and see improvements at high specificity over the CNN especially with location and context features, which contain information not available to the CNN. Additionally, a reader study was performed, where the network was compared to certified screening radiologists on a patch level and we found no significant difference between the network and the readers.
Pub.: 09 Aug '16, Pinned: 09 Jun '17
Abstract: Authors: Deo, R. C ; Nallamothu ; B. K. Article URL: http://circoutcomes.ahajournals.org/cgi/content/short/CIRCOUTCOMES.116.003308v1?rss=1 Citation: (2016) Publication Date: 2016-11-08T13:00:52-08:00 Journal: Circulation : Cardiovascular Quality and Outcomes
Pub.: 08 Nov '16, Pinned: 09 Jun '17
Abstract: Background— Using electronic health records data to predict events and onset of diseases is increasingly common. Relatively little is known, although, about the tradeoffs between data requirements and model utility. Methods and Results— We examined the performance of machine learning models trained to detect prediagnostic heart failure in primary care patients using longitudinal electronic health records data. Model performance was assessed in relation to data requirements defined by the prediction window length (time before clinical diagnosis), the observation window length (duration of observation before prediction window), the number of different data domains (data diversity), the number of patient records in the training data set (data quantity), and the density of patient encounters (data density). A total of 1684 incident heart failure cases and 13 525 sex, age-category, and clinic matched controls were used for modeling. Model performance improved as (1) the prediction window length decreases, especially when
Pub.: 08 Nov '16, Pinned: 09 Jun '17