{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# ARTIFICIAL INTELLIGENCE IN MEDICINE                          \n          \n## INTRODUCTION \n### 1.0 BACKGROUND \nThe world is changing. The interest in a shift to a more developed world has caused an increase in the development and usage of our digital space. Several works and occupations are now dissolved and rendered to the hands of technology and Artificial intelligence (AI).\nThis is also evident in our medical sector. Proofs of a swift and less stressful technique towards tests and diagnosis has led to a concomitant spike in the field of digital pathology. Another contributing factor is rising concerns over uncertain diagnosis and prolonged test result delivery when using traditional pathology. Statistics and societal acceptance of technology has clearly shown that artificial intelligence is here to stay. \nIn that light, this article tends to delve into the contributions of artificial intelligence in medicine, using biomedical sciences as an outlook.\n### 1.1 BIOMEDICAL SCIENCES\nAccording to Wikipedia, Biomedical sciences are a set of sciences applying portions of natural science or formal science, or both, to develop knowledge, interventions, or technology that are of use in healthcare or public health. \nTraditionally, biomedical sciences includes medical microbiology, clinical chemistry, histopathology and genetic epidemiology. With the infusion of artificial intelligence, biomedical sciences also consist of biomedical engineering. \nOne misconception of most people today is that artificial intelligence is an anomaly that emerged recently. Meanwhile that is not the case. It dates back to the 1950s. As old as artificial intelligence is, also depicts how old it's impact to medicine is.\n### 1.2 ORIGIN OF ARTIFICIAL INTELLIGENCE IN MEDICINE\nArtificial intelligence was first coined by John McCarthy in a proposal for a computer science conference that was held at Dartmouth College in 1956.  In the years that followed that conference, several universities developed research programs in AI, often with support from the Defense Advanced Research Projects Agency (DARPA) in the US Department of Defense (Edward, 2019). \nDuring that period, there was no substantial impact of AI to medicine until the late 1960s, where it was incorporated to tackle issues in natural sciences. The first assay to ever be carried out was the Heuristic Dendral project; a program for generating explanative hypothesis in organic chemistry. It was performed by geneticist Joshua Lederberg,  computer scientist Edward Feigenbaum, organic chemist Carl Djerassi and philosopher-of-science Bruce Buchanan. They also applied clever algorithms that inferred likely chemical structures of organic compounds from mass-spectral data. An illustration of such algorithm is represented below. \n![Composition: C8 H16 O](https://ibb.co/VH0BkZk)\nAnother technique invented for problem solving processes in medicine was the Neural network analysis. This belongs to the machine learning (ML) techniques which may be used to extract knowledge from patients to optimize decision limits and identify clinically important laboratory quantities (Ann, 1994).\nThrough neural network analysis came the idea to develop automations which is the forefront of most technological devices in medical pathology. \n### 1.3 RECENT ADVANCES OF ARTIFICIAL INTELLIGENCE IN MEDICINE \nThe recent advances of AI in medicine can be seen in data and information control systems such as electronic medical records, electronic health records and other related medical data. It can also be seen in robotics, which assist surgeons in drug delivery systems and in our dear automations.\nThe standard use for machine learning is to make a prediction based on something that can be measured. For example, in psychiatric medicine, studies have used smartphone recordings of everyday behaviors (e.g., when patients wake up or how much they exercise) to predict mood using machine learning (Wang et al. 2014). \nMachine learning can also provide an objective opinion to improve productivity, consistency, and accurateness. Google has developed a machine learning technique to help recognize cancerous tumors on mammograms (De Gruyter, 2020).\nIn biomedical sciences, it can be seen in whole slide imaging, digital imaging, genomic imprinting and lots more. The table below shows the various advances in biomedical sciences from manual pathology to digital pathology.\n\n| SN | BIOMEDICAL FIELDS | MANUAL METHODS | DIGITAL METHODS |\n|----|-------------------|----------------|-----------------|\n|1.\t|Histopathology |\tStaining a specimen using slides.Manual processing of tissues. |Fully automated staining systems. Tissue processor to automatically prepare tissue samples.|\n|2.\t| Genomic medicine |The use of genetic test kits.|The use of biomarkers and computational and statistical methods to decode the functional information hidden in DNA sequence.|\n|3.\t| Clinical chemistry | The use of test tubes and containers to perform analysis. The use of manual pipette.\t| The use of machines such as Senser core, Caretium and SFRI for analysis. The use of electronic pipette.|\n|4. | Medical microbiology | Culturing of media using petric dishes. Reading a cultured media with microscope. | Automated blood culture devices and automated susceptibility testing instruments. Digital plate reading.|\n|5.\t| Haematology | Checking of blood groups using test tubes or plain tiles. | Using AI-based algorithms to analyze images of blood samples to identify and classify different types of blood cells.|\n### 1.4 CONCLUSION\nIt is clear that AI is the future. It's impact cannot be buttressed enough. Years to come, machine learning will even provide benefits to the family physician at home. Healthcare needs to interchange from intelligence of ML as an innovative perception to sight it as a real-world tool that can be organized nowadays (Gruyter, 2020). \nWe must take an incremental approach if ML has to play a role in healthcare system. Prospective medical personnels and technophobic medics needs to be sensitized and enlightened on the benefits of artificial intelligence. \nRealistically it would take years for machine learning to be accepted globally, but with little efforts, one candle lighting a thousands, and even with the insights I have shared on this article, our medical space would one day be completely automated.\n### 1.5 REFERENCES\nAnn Biol Clin (1994), Artificial intelligence within the chemical laboratory.\n\nDe Gruyter(2020),  Recent advancement of machine learning and deep learning in the field of healthcare system. https://www.degruyter.com/search?query=*&publisherFacet=De+Gruyter\n\nEdward H. Shortlife (2019), Artificial intelligence in medicine: Weighing the accomplishments, hype and promise. \n\nWang R, Chen F, Chen Z, Li T, Harari G, Tignor S, Zhou X, Ben-Zeev D, Campbell AT. (2014). Assessing mental health, academic performance and behavioral trends of college students using smartphones.","metadata":{}}]}