Author Interviews, Genetic Research, Nature, Technology / 15.07.2020 Interview with: Dr.Altuna Akalin PhD Head of Bioinformatics and Omics Data Science Platform Berlin Institute for Medical Systems Biology (BIMSB) Max Delbrück Center for Molecular Medicine (MDC) Berlin, Germany What is the background for this study? Where does the word Janggu come from?  Response: Deep learning applications on genomic datasets used to be a cumbersome process where researchers spend a lot of time on preparing and formatting data before they even can run deep learning models. In addition, the evaluation of deep learning models and the choice of deep learning framework were also not straightforward. To streamline these processes, we developed JangguWith this framework, we are aiming to relieve some of that technical burden and make deep learning accessible to as many people as possible. Janggu is named after a traditional Korean drum shaped like an hourglass turned on its side. The two large sections of the hourglass represent the areas Janggu is focused: pre-processing of genomics data, results visualization and model evaluation. The narrow connector in the middle represents a placeholder for any type of deep learning model researchers wish to use.  (more…)
Author Interviews, Dermatology, Technology / 03.04.2020 Interview with: Jung-Im Na, MD PhD Associate Professor, Department of Dermatology Seoul National University Bundang Hospital Korea What is the background for this study? Would you briefly explain what is meant by a convolutional neural network? Response: When a very young child looks at a picture, she can easily identify cats and dogs, however, even the most advanced computers had struggled at this task until recently. Computers began to “see” with the recent advancement of Deep Learning techniques. Deep Learning is a machine learning technique that teaches computers to learn from raw data. Most deep learning methods use artificial neural network architectures, imitating human brain, and convolutional neural networks (CNN) is a particular type of deep learning architecture, imitating the visual cortex. CNN is especially powerful for recognizing images. CNN exploit the information contained in image datasets to automatically learn features and patterns. (more…)
Author Interviews, Columbia, COVID -19 Coronavirus, NYU, Technology / 02.04.2020 Interview with: Professor Anasse Bari PhD Courant Institute of Mathematical Sciences, Computer Science Department, New York University, New York, and Megan Coffee MD PhD Division of Infectious Diseases and Immunology, Department of Medicine New York University, Department of Population and Family Health Mailman School of Public Health Columbia University, New York What is the background for this study? Coffee and Bari:  This work is led by NYU Grossman School of Medicine and NYU’s Courant Institute of Mathematical Sciences, in partnership with Wenzhou Central Hospital and Cangnan People's Hospital, both in Wenzhou, China. This is a multi-disciplinary team with backgrounds in clinical infectious disease as well as artificial intelligence (AI) and computer science. There is a critical need to better understand COVID-19. Doctors learn from collective and individual clinical experiences. Here, no clinician has years of experience. All are learning as they go, having to make important decisions about clinical management with stretched resources. The goal here is to augment clinical learning with machine learning. In particular, the goal is to allow clinicians to identify early who from the many infected will need close medical attention. Most patients will first develop mild symptoms, yet some 5-8 days later will develop critical illness. It is hard to know who these people are who will need to be admitted and may need to be intubated until they become ill. Knowing this earlier would allow more attention and resources to be spent on those patients with worse prognoses. If there were ever treatments in the future that could be used early in the course of illness, it would be important to identify who would most benefit We present in this study a first step in building an artificial intelligence (AI) framework, with predictive analytics (PA) capabilities applied to real patient data, to provide rapid clinical decision-making support. It is at this point a proof of concept that it could be possible to identify future severity based on initial presentation in COVID-19. (more…)
Author Interviews, Dermatology, JAMA, Melanoma, Technology / 21.10.2019 Interview with: Dr. Helen Marsden PhD Skin Analytics Limited London, United Kingdom What is the background for this study? Response: In this technology age, with the explosion of interest and applications using Artificial Intelligence, it is easy to accept the output of a technology-based test - such as a smartphone app designed to identify skin cancer - without thinking too much about it. In reality, technology is only as good as the way it has been developed, tested and validated. In particular, AI algorithms are prone to a lack of “generalisation” - i.e. their performance drops when presented with data it has not seen before. In the medical field, and particularly in areas where AI is being developed to direct a patient’s diagnosis or care, this is particularly problematic. Inappropriate diagnosis or advice to patients can lead to false reassurance, heightened concern and pressure on NHS services, or worse. It is concerning, therefore, that there are a large number of smartphone apps available that provide an assessment of skin lesions, including some that provide an estimate of the probability of malignancy, that have not been assessed for diagnostic accuracy. Skin Analytics has developed an AI-based algorithm, named: Deep Ensemble for Recognition of Malignancy (DERM), for use as a decision support tool for healthcare providers. DERM determines the likelihood of skin cancer from dermoscopic images of skin lesions. It was developed using deep learning techniques that identify and assess features of these lesions which are associated with melanoma, using over 7,000 archived dermoscopic images. Using these images, it was shown to identify melanoma with similar accuracy to specialist physicians. However, to prove the algorithm could be used in a real life clinical setting, Skin Analytics set out to conduct a clinical validation study. (more…)
Author Interviews, NYU, PTSD, Technology / 22.04.2019 Interview with: Charles R. Marmar, MD The Lucius N. Littauer Professor Chair of the Department of Psychiatry NYU Langone School of Medicine What is the background for this study? What are the main findings?  Response: Several studies in recent years have attempted to identify biological markers that distinguish individuals with PTSD, with candidate markers including changes in brain cell networks, genetics, neurochemistry, immune functioning, and psychophysiology. Despite such advances, the use of biomarkers for diagnosing PTSD remained elusive going into the current study, and no physical marker was applied in the clinic. Our study is the first to compare speech in an age and gender matched sample of a military population with and without PTSD, in which PTSD was assessed by a clinician, and in which all patients did not have a major depressive disorder. Because measuring voice qualities in non-invasive, inexpensive and might be done over the phone, many labs have sought to design speech-based diagnostic tools  (more…)
Author Interviews, Lung Cancer, Nature, NYU, Technology / 17.09.2018 Interview with: Aristotelis Tsirigos, Ph.D. Associate Professor of Pathology Director, Applied Bioinformatics Laboratories New York University School of Medicine What is the background for this study? What are the main findings? Response: Pathologists routinely examine slides made from tumor samples to diagnose cancer types. We studied whether an AI algorithm can achieve the same task with high accuracy. Indeed, we show that such an algorithm can achieve an accuracy of ~97%, slightly better than individual pathologists. In addition, we demonstrated that AI can be used to predict genes that are mutated in these tumors, a task that pathologists cannot do. Although the accuracy for some genes is as high as 86%, there is still room for improvement. This will come from collecting more training data and also from improvement in the annotations of the slides by expert pathologists.   (more…)
Author Interviews, Schizophrenia, Technology / 13.07.2018 Interview with: Bo Cao, Ph.D. Assistant Professor Department of Psychiatry Faculty of Medicine & Dentistry University of Alberta Edmonton What is the background for this study? What are the main findings? Response: Schizophrenia is a severe psychiatric disorder that comes with delusions, hallucinations, poor motivation, cognitive impairments. The economic burden of schizophrenia was estimated at $155.7 billion in 2013 alone in the United States. Schizophrenia usually emerges early in life and can potentially become a lifetime burden for some patients. Repeated untreated psychotic episodes may be associated with irreversible alterations of the brain. Thus, it is crucial to identify schizophrenia early and provide effective treatment. However, identifying biomarkers in schizophrenia during the first episode without the confounding effects of treatment has been challenging. Limited progress has been made in leveraging these biomarkers to establish diagnosis and make individualized predictions of future treatment responses to antipsychotics. In a recent study by Dr. Cao and his colleagues, they successfully identified the first-episode drug-naïve schizophrenia patients (accuracy 78.6%) and predict their responses to antipsychotic treatment (accuracy 82.5%) at an individual level by using a machine learning algorithm and the functional connections of a brain region called the superior temporal cortex.  (more…)
ASCO, Author Interviews, Cancer Research, Technology / 22.06.2018 Interview with: Alexandra Urman, MPH Clinical Research Manager Clinical Development IBM Watson Health What is the background for this study?  Response: Cancer statistics show only 3-5% of cancer patients participate in clinical trials although up to 20% may be eligible. Dr. Tufia Hadad, a medical Oncologist at the Mayo Clinic in Rochester, Minnesota sought to address this issue and spearheaded a project conducted at the Rochester facility in collaboration with IBM Watson Health. The objective was to determine if the use of cognitive computing increased clinical trial enrollment and screening efficiency in the breast cancer clinic. Watson for Clinical Trial Matching (CTM) is a cognitive system which utilizes natural language processing to derive patient and tumor attributes from unstructured text in the electronic health record that can be further used to match a patient to complex eligibility criteria in trial protocols. (more…)
Author Interviews, Technology / 21.01.2018 Interview with: Jacob Crandall PhD Associate Professsor, Computer Science Brigham Young University What is the background for this study? Response: As autonomous machines become increasingly prevalent in society, they must have the ability to forge cooperative relationships with people who do not share all of their preferences.  Unlike the zero-sum scenarios (e.g., Checkers, Chess, Go) often addressed by artificial intelligence, cooperation does not require sheer computational power.  Instead, it is facilitated by intuition, emotions, signals, cultural norms, and pre-evolved dispositions.  To understand how to create machines that cooperate with people, we developed an algorithm (called S#) that combines a state-of-the-art reinforcement learning algorithm with mechanisms for signals. We compared the performance of S# with people in a variety of repeated games. (more…)
Author Interviews, Stem Cells, Technology / 14.12.2017 Interview with: Osipov PhD Insilico Medicine and Dmitry Klokov PhD Canadian Nuclear Laboratories, Chalk River, Ontario, Canada What is the background for this study? What are the main findings? Response: Cells and tissues can be damaged when exposed to ionizing radiation. In case of radiotherapy, it is a desirable effect in tumor cells. In case of occupational, medical and accidental exposures, typically to low-dose radiation, this may pose health risk to normal cells and tissues. In both cases, short-term assays that quantify damage to DNA and help evaluate long-term outcome are key to treatment/risk management. One such short-term assay is based on quantification of a modified histone protein called gH2AX in exposed cells up to 24 hrs after exposure. This protein marks sites in DNA that have both strands of the DNA helix broken or damaged. This assay is also widely used for various applications, including determination of individual radiosensitivity, tumor response to radiotherapy and biological dosimetry. With the advent of regenerative medicine that is based on stem cell transplantation, the medical and research communities realized that there is a need to understand how stem cells respond to low-dose diagnostic radiation exposures, such as CT scans. Stem cell therapies may have to be combined with diagnostic imaging in recipient patients. The gH2AX assay comes in very handy here, or at least it seemed this way. We exposed mesenchymal stem cells isolated from human patients to low or intermediate doses of X-rays (80 and 1000 mGy) and followed formation of gH2AX in their nuclei. First we found that residual gH2AX signal in cells exposed to a low dose was higher than in control non-irradiated cells. If the conventional assumptions about this assay that it is a surrogate for long-term detrimental effects was followed it would mean that the low-dose exposed cells were at a high risk of losing their functional properties. So we continued growing these cells for several weeks and assayed gH2AX levels, ability to proliferate and the level of cellular aging. Surprisingly, we found that low-dose irradiated cells did not differ from non-irradiated cells in any of the measured functional end-points. This was in contrast to 1000 mGy irradiated cells that did much worse at those long-term end points. (more…)
Author Interviews, Breast Cancer, Cancer Research, JAMA, Technology / 13.12.2017 Interview with: Babak Ehteshami Bejnordi Department of Radiology and Nuclear Medicine Radboud University medical center, NijmegenBabak Ehteshami Bejnordi Department of Radiology and Nuclear Medicine Radboud University medical center, Nijmegen What is the background for this study? Response: Artificial intelligence (AI) will play a crucial role in health care. Advances in a family of AI popularly known as deep learning have ignited a new wave of algorithms and tools that read medical images for diagnosis. Analysis of digital pathology images is an important application of deep learning but requires evaluation for diagnostic performance. Accurate breast cancer staging is an essential task performed by the pathologists worldwide to inform clinical management. Assessing the extent of cancer spread by histopathological analysis of sentinel lymph nodes (SLN) is an important part of breast cancer staging. Traditionally, pathologists endure time and labor-intensive processes to assess tissues by reviewing thousands to millions of cells under a microscope. Using computer algorithms to analyze digital pathology images could potentially improve the accuracy and efficiency of pathologists. In our study, we evaluated the performance of deep learning algorithms at detecting metastases in lymph nodes of patients with breast cancer and compared it to pathologist’s diagnoses in a diagnostic setting. (more…)
Author Interviews, Technology / 22.04.2017 Interview with: Aylin Caliskan PhD Center for Information Technology Policy Princeton University, Princeton, NJ What is the background for this study? What are the main findings? Response: Researchers have been suggesting that artificial intelligence (AI) learns stereotypes, contrary to the common belief that AI is neutral and objective. We present the first systematic study that quantifies cultural bias embedded in AI models, namely word embeddings. Word embeddings are dictionaries for machines to understand language where each word in a language is represented by a 300 dimensional numeric vector. The geometric relations of words in this 300 dimensional space make it possible to reason about the semantics and grammatical properties of words. Word embeddings represent the semantic space by analyzing the co-occurrences and frequencies of words from billions of sentences collected from the Web. By investigating the associations of words in this semantic space, we are able to quantify how language reflects cultural bias and also facts about the world. (more…)
Author Interviews, Cancer Research, Columbia, Genetic Research, Personalized Medicine / 23.12.2016 Interview with: Dr. Kai Wang Zilkha Neurogenetic Institute, University of Southern California Institute for Genomic Medicine, Columbia University What is the background for this study? What are the main findings? Response: Cancer is a genetic disease caused by a small number of “driver mutations” in the cancer genome that drive disease initiation and progression. To understand such mechanism, there are increasing community efforts in interrogating cancer genomes, transcriptomes and proteomes by high-throughput technologies, generating huge amounts of data. For example, The Cancer Genome Atlas (TCGA) project has already made public 2.5 petabytes of data describing tumor and normal tissues from more than 11,000 patients. We were interested in using such publicly available genomics data to predict cancer driver genes/variants for individual patients, and design an "electronic brain” called iCAGES that learns from such information to provide personalized cancer diagnosis and treatment. iCAGES is composed of three consecutive layers, to infer driver variants, driver genes and drug treatment, respectively. Unlike most other existing tools that infer driver genes from a cohort of patients with similar cancer, iCAGES attempts to predict drivers for individual patient based on his/her genomic profile. What we have found is that iCAGES outperforms other tools in identifying driver variants and driver genes for individual patients. More importantly, a retrospective analysis on TCGA data shows that iCAGES predicts whether patients respond to drug treatment and predicts long-term survival. For example, we analyzed two groups of patients and found that using iCAGES recommend drugs can increase patients’ survival probability by 66%. These results suggest that whole-genome information, together with transcriptome and proteome information, may benefit patients in getting optimal and precise treatment. (more…)