Readings Weekend 1 HCDA 550
Among adults with chronic illness, ____% to ____% of medications are not taken as prescribed. In the United States, it is estimated that medication _______ is associated with 125 000 deaths, 10% of hospitalizations, and $100 billion in health care services annually.
30, 50, nonadherence
Classification or regression technique that uses a multitude of models to come up with a decision but weighs them based on their accuracy in predicting the outcome is called ___ ___ and is a type of ______ learning
Adaboost, supervised
future winners for genetic testing will harness _____ technology
EMR
Flatiron harnesses NGS to works towards improving precision medicine in oncology along with a handful of other techniques like _______, longitudinal clinical data, claims billing
EMRs
A multilayered neural network with a special architecture designed to extract increasingly complex features of the data at each layer to determine the output
convolutional neural network
if you have an unstructured data set (e.g. images) and you need to infer information from it, you would use
convolutional neural network
In the _______ vertical model, the value of an invention is pursued as far as possible by the inventor, as opposed to selling the idea off before its full value is realized.
Fully integrated biopharmaceutical company
A generalization of k-means clustering that provides more flexibility in the size and shape of groups (clusters) called ___ ___ and is a type of ______ learning
Gaussian mixture model, unsupervised
an interdisciplinary field of science focusing on the structure, function, evolution, mapping, and editing of genomes. personalized medicine
Genomics
Puts data into a number of groups (k) that each contain data with similar characteristics (as determined by the model, not in advance by humans) is called ___ ___ and is a type of ______ learning
K-means clustering, unsupervised
two common types of deep learning:
convolutional neural network, recurrent neural network
a company in-licenses product from others that is already in preclinical or clinical testing. An example is The Medicines Company; that firm does not engage in drug discovery.
NRDO
What is replacing the Sanger method in sequencing DNA?
New generation sequencing (NGS)
Blockchain is still three to five years away from feasibility at ______, primarily because of the difficulty of resolving the "_______" paradox to establish ____ _____.
scale, coopetition, common standards
Recent studies have demonstrated the survival benefits of recording _________. Delivering these insights back to care teams in a timely manner enables precision intervention. Important for health app integration
PROs
focus on the earlier stages in the value chain, such as discovery and preclinical development. The model covers platform and tool-based companies seeking to commercialise drug targets, services and technologies that can be sold or licensed to other companies.
RIPCOs
The term________________ was coined to describe a cellular mechanism that use the gene's own DNA sequence of gene to turn it off, a process that researchers call silencing. siRNAs (silent RNAs) can be designed to match any gene, can be manufactured cheaply, and can be readily administered to cells. One can now order commercially synthesized siRNAs to silence virtually any gene in a human or other organism's cell, dramatically accelerating the pace of biomedical research. Furthermore, the ability to turn off expression of a single gene makes RNAi an appealing therapeutic approach to treat infectious diseases or genetic disorders, such as those that result from the inappropriate and undesirable activity of a gene, as in many cancers and neurodegenerative diseases. There are currently several clinical trials testing the safety and effectiveness of siRNA drugs.
RNA interference
PhrMA companies focus on one technology which is cutting edge and applicable to . multiple disease areas
Technology platform model
Oncology has focused on a _______; however, it is confusing how this should be implemented some countries use ______, yet value varies by _________ (patients, caregivers, payers, and providers are misaligned in their values) oncology spending places a large strain on healthcare system, so we need to refine these _____
VAF, qalys, individual, VAFS
companies may outsource/contract extensively for services at any point(s) in the value chain, providing access to complementary assets outside the firm. This allows a company to maintain control of the product development process and defer the point at which they plug into the value chain.
Virtual Model
Blockchain has the potential to streamline and accelerate R&D, boosting innovation and lowering costs at pharma and medtech companies. For example, it can automatically validate, coordinate, and maintain consistent adherence to approved trial protocols across R&D projects and sites. Patients can self- identify and enroll in clinical trials, and the system will automatically assess and verify their eligibility. Researchers can quickly gather needed clinical data in reliably verified ways, as the system stores all data in a consistent, accessible infrastructure in which patients grant access to others by sharing public and private keys is an example of _____ ______ using blockchain technology
accelerating R&D
Companies can makes lots of money by ________ _____ ______ (see 23andMe model)
aggregating genetic information
Quantum ____________ are the tools that tell quantum computers what to do. Two of their attributes are especially important in the near term: 1. _______ (how much faster can it solve the problem that traditional tech 2. ________ (How resilient is the algo- rithm to the random "noise," or other errors, in quantum computing?)
algorithms, speed-up, robustness
_______ technology presents an opportunity to manipulate gene expression within cells to treat an endless number of diseases and is a powerful tool for studying gene function. The antisense approach utilizes antisense agents to fight various diseases by regulating the expression of a specific factor, the presence of which actually causes the particular disease. Highly specific and effective gene silencing of any disease can be achieved by an accurate knowledge of the target mRNA sequence and rational design of its complementary antisense agents for the downregulation of its protein message. This technology has been successfully used for the treatment of cancer, HIV, and other mutating viral diseases.
antisense
typically defined as the ability of a machine to perform cognitive functions we associate with human minds, such as perceiving, reasoning, learning, and problem solving
artificial intelligence
_______ in blockchain are often new market entrants without an existing market share to protect, so they need to seek disruptive or transformative business models and blockchain solutions. Attacker approaches are suited to use cases with the highest disruptive potential through offering a service to the market that would disintermediate existing players. _____________ _________ often adopt an attack strategy because they are selling the services into—and disrupting—industries in which they are not currently participants.
attackers, blockchain as a service (BaaS) providers
A ________ is a distributed online ledger that can prove whether a piece of data is __________
blockchain, authentic
_______ in blockchain: need to be driving the conversations and consortiums that are shaping the new standards that will disrupt their current businesses. Despite being dominant players, they cannot single handedly direct blockchain adoption as they face greater ______ and standardization barriers. Instead, they can position themselves to shape and capture the value of new blockchain standards.
conveners, regulatory
healthcare and blockchain?
can initially reduce costs
the alterations in cellular metabolism pathways that are evident in cancer cells compared with most normal tissue cells. Metabolic alterations in cancer cells are numerous and include aerobic glycolysis, reduced oxidative phosphorylation and the increased generations of biosynthetic intermediates needed for cell growth and proliferation
cancer metabolism
These trials suggest that that clinicians may improve adherence by educating patients about their personalized risks with and without therapy at the time of disease diagnosis and by conducting follow-up telephone calls (using clinic support staff if avail- able) to reinforce risks of nonadherence. However, providing timely and sustained interventions within the clinical workflow may be ____________ for many clinicians.
challenging
Photons as quantum computing technology: pros: existing technology for _____ design; operates at _____ _________ con: controlling ______ interactions (key for two-qubit gates)
chip, room temperature, multiphoton
Problems with _____ in blockchain: There are also significant barriers to collaboration. Achieving blockchain's full potential in health care requires the adoption of common data definitions, an issue that has bedeviled the industry for years.
collaboration
medication mamngement regumen: Simplifying medication regimens by maximizing doses of 1 medication before starting another and using _____ pills when available can improve adherence.
combination
laboratory technique in which millions of molecular constructions can be synthesized and tested for biological activity. It has generated massive numbers of targeted molecules for testing and the developing techniques of high throughput screening has automated the screening process so larger numbers of biological evals can be done
combinatorial chemistry
main problems with blockchain's feasibility at scale: ____ _______ _____ must be essential _____ must advance (think estonia's off-chain records) _____ must be digitized the ______ paradox must be resolved
common regulatory structures, technology, assets, coopetition
Despite the association between electronically monitored adherence and improved clinical outcomes, the ______ of the monitors and the challenges of ______ their adherence data into clinical care may be barriers to their routine use outside of the research setting.
cost, integrating
Six levers to unlock machine-learning potential in healthcare 1. _____ ______ is critical, yet often overlooked 2. Results must be _______ 3. Algorithms must be ______ 4. Algorithms must be _____ 5. Algorithms must demonstrate _____ 4. Algorithms must be _____
data quality, reproducible, transparent, credible, impact, fair
Blockchain's core advantages are ______, cryptographic _____, __________, and _______
decentralization, security, transparency, immutability
Highly interpretable classification or regression model that splits data-feature values into branches at decision nodes (eg, if a feature is a color, each possible color becomes a new branch) until a final decision output is made is called ___ ___ and is a type of ______ learning
decision tree, supervised
a type of machine learning that can process a wider range of data resources, requires less data preprocessing by humans, and can often produce more accurate results than traditional machine-learning approaches.
deep learning
which is better? deep learning or other traditional forms of learning (like machine learning)
deep learning
describe what happened employed across all industries
descriptive analytics
Health apps: The ________ era creates an opportunity to more closely align multistakeholder values with the _________ at the center. This in turn helps future __________ incorporate the latest elements of precision care delivery for the value era in oncology.
digital, patient, VAFs
Blockchain does not have to be a ________ to generate value
disintermediator
_________ players who can establish their blockchains as the market solutions should make big bets now.
dominant
A type of protein made in the laboratory that can bind to substances in the body, including cancer cells. There are many kinds. They are made so that it binds to only one substance. They can be used alone or to carry drugs, toxins, or radioactive substances directly to cancer cells.
monoclonal antibody
Types of interventions to improve adherence: 1. patient ______ 2. Medication ____ ______ 3. clinical _____ _______ 4. ____ ______ _____ 5. medication-taking ______
education, regimen management, pharmacist consultation, cognitive behavioral therapy, reminders
However, adherence measured by pharmacy claims may not be as strongly predictive of clinical outcomes compared with _____ _______ _________, since pharmacy claims may not necessarily reflect the medications patients actually ingest.
electronic pill bottles
4 categories of quantum computer providers: 1. ___________ (big tech companies (IBM) and well-funded startups) 2. _________ (Can expand to software, etc., tend to be tech giants) 3. _________ (confusing group, partner with hardware providers often) 4. ________ (mainly startups that provide solutions to many quantum computing issues) The ecosystem is dynamic, and the lines between tech layers are easily blurred.
end-to-end providers, hardware and systems players, software and service players,
the study of biological mechanisms that will switch genes on and off. DNA gives the instructions for various functional proteins to be produced inside the cell - this process is also known as the central dogma of molecular biology.
epigenetics
______ in blockchain: should also carefully consider and implement an appropriate blockchain strategy. Most companies do not have the capability to influence all necessary parties, especially when applications of blockchain require high standardization or regulatory approval. Such companies cannot be unaware of market innovations—they should keep a watching brief on blockchain developments and be prepared to move fast to adopt emerging standards.
followers
Problems with _____ in blockchain: Before patients, providers, and others feel entirely comfortable handing over personal data to a blockchain, we expect that laws and regulations—including laws governing blockchain's smart contracts, procedures for data sharing, and reporting of financial transactions—will have to supersede technological self-enforcement. Many tricky issues remain unsolved.
funding
issues with ________ in blockchain: Compared with investments in artificial intelligence for health care, which totaled almost $800 million in 2016, the amounts invested in blockchain thus far are small. The pace of funding activity is accelerating, however. Large players such as IBM and Microsoft are bullish about incorporating blockchain platforms and development in their customer support systems.
funding
Blockchain will not disrupt health care overnight—or even over the next few years. Various aspects of _____, ______ and ______, ______, and ______ need to develop further before we reach a tipping point in adoption and impact. Even so, lots of players are aggressively pursuing solutions.
funding, technology, talent, governance, collaboration
Classification technique that applies Bayes theorem, which allows the probability of an event to be calculated based on knowledge of factors that might affect that event (eg, if an email contains the word "money," then the probability of it being spam is high)is called ___ ___ and is a type of ______ learning
naive bayes, supervised
______ _____-involves altering the genes inside your body's cells in an effort to treat or stop disease. Scientists are discovering various ways to perform gene therapy, including: o ______ mutated genes: some cells become diseased because certain genes work incorrectly or no longer work at all. Replacing the defective genes may help treat certain diseases. For instance, a gene called p53 normally prevents tumor growth. Several types of cancer have been linked to problems with the p53 gene. If doctors could replace the defective p53 gene, that might trigger the cancer cells to die. o ______ mutated genes: Mutated genes that cause disease could be turned off so that they no longer promote disease, or healthy genes that help prevent disease could be turned on so that they could inhibit the disease. o Making diseased cells more evident to the _______ system: In some cases, your immune system doesn't attack diseased cells because it doesn't recognize them as intruders. Doctors could use gene therapy to train your immune system to recognize the cells that are a threat.
gene therapy, replacing, fixing, immune
Classification or regression technique that generates decision trees sequentially, where each tree focuses on correcting the errors coming from the previous tree model. The final output is a combination of the results from all trees is called ___ ___ and is a type of ______ learning
gradient-boosting tree, supervised
3 layers of quantum computing: quantum computer _______ (qubits are built here) control ______ (responsible for quantum gates, error correcting, etc.) _____ (implements algorithms/ applications, etc.)
hardware, systems, software
______—specifically, the understanding of diseases and treatments—is fundamentally different from other areas where machine learning has been used.
health
Splits or aggregates clusters along a hierarchical tree to form a classification system is called ___ ___ and is a type of ______ learning
hierarchical clustering, unsupervised
blockchain: private/ permissionless=
high scalability
recent scientific methods relevant to the field of chemistry and biology, in which hundreds of thousands of experimental samples are subjected to simultaneous testing under given conditions. The sample themselves may take the form of biochemical agents such as chemical compounds, amino acids, or live cells. With the development of laboratory robotics that automate sample preparation, handling and data analysis, scientists can easily and reliably generate and use large datasets from these HTS to answer complex biological questions. HTS is now widely used in the field of pharmaceuticals, biotech and academic institutes for drug discovery, target validation and the identification of genes or proteins that modulate a particular biological pathway.
high throughput screening
Three major trends that will affect the market for genetic testing: 1. data _____ and ______ (helped by comprehensive EMRS, but must abide by data privacy laws) 2. increasing ______ pressures (healthcare spending is increasing faster than insurance premiums, therefore payers will harness genetic information to establish ______ treatments to mitigate costs) 3. evolving US _____ landscape (high reimbursement rates currently for genetic testing, especially by _____)
integration, analysis, cost, preventative, reimbursement, CMS
Adherence can be assessed and improved within the context of usual clinical care, but more ____ and ______ interventions that have demonstrated success will require additional investments by health systems.
intensive, costly
quantum computing technology: The core elements are single ions (charged atoms) trapped in magnetic fields and the energy levels of their intrinsic spin form the qubit.
ion traps
Silicon-based cubit technologies as quantum computing technology: pros: research has made significant strides cons: still must master ________ in order to implement these systems
nanoegineering
For many years, the biggest potential end users for quantum computing capability were _____ _____ not so much anymore bexause of uncertainties of applicability
national governments
Two biggest questions face emerging quantum computing industry: When will we have a ____, _____ quantum computer? What will be its _____?
large, reliable, architecture
This consideration of a company's market position will inform which of four distinct strategic approaches to blockchain should be deployed and, in fact, further refine which type of use cases to focus on first. four types:
leaders, conveners, followers, attackers
_________ in blockchain: The greatest risk for these companies is ________, which would cause them to lose the opportunity to strengthen their competitive advantages compared to competitors. An example of a leader following this strategy is Change Healthcare, one of the largest independent healthcare IT companies in the United States when it launched an enterprise- scale healthcare blockchain for claims processing and payment.
leaders, inaction
Highly interpretable, standard method for modeling the past relationship between independent input variables and dependent output variables (which can have an infinite number of values) to help predict future values of the output variables is called ___ ___ and is a type of ______ learning
linear regression, supervised
Upgrades a logistic regression to deal with nonlinear problems—those in which changes to the value of input variables do not result in proportional changes to the output variables is called ___ ___ and is a type of ______ learning
linear/ quadratic regression analysis, supervosed
Extension of linear regression that's used for classification tasks, meaning the output variable is binary (eg, only black or white) rather than continuous (eg, an infinite list of potential colors) is called ___ ___ and is a type of ______ learning
logistic regression, supervised
blockchain: public/ permissionless=
low scalability
algorithms detect patterns and learn how to make predictions and recommendations by processing data and experiences, rather than by receiving explicit programming instruction.
machine learning
blockchain: public/ permissioned=
medium scalability
Which are big businesses investing in superconducting cubits most? 1. based on _____ technology (CMOS).. somewhat standardized and easier to handle 2. near-term path to _____ _______ is more in sight because can support over 50 ______ unlike ion traps
medium scalability, qubits
What should ____ do about blockchain? Being able to enter into contracts with patients about the exchange and use of data could help overcome constraints on innovation that HIPAA and other privacy regulations have imposed. Medtech developers will want to explore how they can use blockchain technology to secure data transactions among connected devices (especially remote devices such as pacemakers, insulin pumps, and sleep apnea equipment) and perhaps to establish broader ecosystems (patient, provider, medtech supplier) around connected devices.
medtech developers
is a tool used to determine whether the DNA from a particular individual contains a mutation in genes like BRCA1 and BRCA2. The chip consists of a small glass plate encased in plastic. Some companies manufacture microarrays using methods similar to those used to make computer microchips. On the surface, each chip contains thousands of short, synthetic, single-stranded DNA sequences, which together add up to the normal gene in question, and to variants (mutations) of that gene that have been found in the human population. Today, DNA microarrays are used in clinical diagnostic tests for some diseases. Sometimes they are also used to determine which drugs might be best prescribed for particular individuals, because genes determine how our bodies handle the chemistry related to those drugs. With the advent of new DNA sequencing technologies, some of the tests for which microarrays were used in the past now use DNA sequencing instead. But microarray tests still tend to be less expensive than sequencing, so they may be used for very large studies, as well as for some clinical tests.
microarrays
in deep learning, interconnected layers of software-based calculators known as "____" form a _____ ____________ The network can then make a determination about the data, learn if its determination is correct, and use what it has learned to make determinations about new data. For example, once it learns what an object looks like, it can recognize the object in a new image.
neurons, neural network
do people intend to replace current systems with quantum computing?
no, just add novel approach
Innovations in oncology like _____ chemotherapy have created ______ for patients to maintain their own care
oral, difficulty
While self-reported adherence is moderately correlated with objective measures of adherence, it tends to _______ adherence.
overestimate
two main features of quantum computing: ______ of zeroes and ones qubits are ________ however, also prone to _____ given complexity of ____ ______... can be corrected however with auxilary _______
overlays, entangled, errors, quantum gates, algorithms
__________ is likely to be the most practical method of assessing adherence in the context of clinical care but may _______ adherence. Hospitals and health systems could consider investments in more intensive but costly strategies (such as recurrent counseling sessions with health educators or reduction/elimination of medication co-payments) to improve medication nonadherence.
patient self-report, overestimate
What should ____ do about blockchain? If patients truly controlled their own data, the benefits of blockchain to providers and payers (as well as to patients) in terms of efficiency and improved service capabilities would be difficult to overstate. However, the industry would also be open to an unprecedented risk of disruption. Block- chain technology can accelerate exposure to potential disintermediation, lowering switching costs and making it easier for startup services to enter the market and challenge the dominance of established paradigms of care.
patients and payers
pharmacist consultation: Although not all trials demonstrated consistent clinical improvements, the evidence suggests that clinicians can improve adherence and potentially disease control by referring ____ to clinical _____ for disease co-management, if available within their practice or health system.
patients, pharmacists
two key features that determine the scalability of blockchain
permissions to read/ write/ commit privacy of servers
What should ____ do about blockchain? Supply chains are among the business areas most likely to feel block- chain's influence. Companies can realize significant value by ring-fencing supply chains from counterfeiters, validating product quality, improving operational efficiency, and accelerating R&D.
pharma
Among the 49 trials included in this review, the most common methods of assessing nonadherence were through: _______ claims data on missed medication fills or refills (23 trials), ______ _______ _______ (14 trials) ______ _______ _______ (9 trials)
pharmacy, electric drug monitors, patient self report
Two other promising technologies of quantum computing: ______ ______-based qubit technologies
photons, silicon
three reasons health is fundamentally different from other areas where machine learning has been used: -despite notable gains in the understanding of disease, the most common health conditions remain ______ understood multifactorial processes. -Second, gains from mechanistic drug development have led to the expectation that we fully understand the _______ of __________. -despite the volume of healthcare data generated during current practice, much information that could meaningfully support insights remains uncollected, siloed, or otherwise ________ to researchers.
poorly, impact, interventions, inaccessible
4 ways to approach quantum computing for various players: 1. analyze ______ (quantify it) 2. gain _______ (experiment with quantum computing) 3. lead your own ______ (build your own quantum unit and collaborate) 4. launch ___ _______ (requires collaboration and lots of investment)
potential, experience, effort, new services
Authors of McKinsey report argue that blockchain has potential, but businesses should view it with ____ ______ based on their market power, industry, etc.
pragmatic skepticism
anticipate what will happen employed in data-driven organizations as a key source of insight
predictive analytics
machine learning focuses on _______ and _______ analytics
predictive, prescriptive
provide recommendations on what to do to achieve goals employed heavily by leading data and internet companies
prescriptive analytics
the large-scale study of proteomes. A proteome is a set of proteins produced in an organism, system, or biological context. We may refer to, for instance, the proteome of a species (for example, Homo sapiens) or an organ (for example, the liver). The proteome is not constant; it differs from cell to cell and changes over time. To some degree, the proteome reflects the underlying transcriptome.
proteomics
two types of quantum algorithms: 1. ______—they are built for speed in noiseless or error-corrected environments. (very susceptible to noise) 2. ____________ are very sturdy algorithms, but they have a somewhat uncertain speed-up over classical algorithms. (designed to be robust in the face of noise and errors. They might have built-in error mitigation, but the most important feature is their shallow depth—that is, the number of gate operations is kept low.) Which are being more invested in?
purebred, workhorses, workhorses
One company exploring blockchain's poten- tial with respect to patient data, Google's DeepMind, is working with Britain's Na- tional Health Service (NHS) to deploy a private blockchain (a blockchain distribut- ed within a closed network of participants) to create robust audit trails that track ex- actly what happens to personal data. The goal is to enable the NHS—and eventually patients—to track personal data access and use in real time. is an example of _____ _____ ______ using blockchain technology
putting patients in charge of their data
Despite many announcements, though, we have yet to see an application where ______ ______—that is, performance by a quantum computer that is superior in terms of time, cost, or quality—has been achieved.
quantum advantage
One novel type of quantum computing:
quantum annealer
quantum annealers do not execute ____ _____ like other technologies; rather, it solves _____ problems
quantum gates, optimization
We use three sets of criteria to assess the current status of the leaders pursuing the main- stream circuit-based approaches and the challenges they still need to overcome. (quantum computers) 1. size of the ____ _____ consisting of _____ and quantum ______ 2. ______ of _______ calculations 3. technical maturity and _______
qubits, quantum, complexity, accurate, scalability
Private placement PIPE (private investments, public equity) R&D limited investments and SWORD Convertible debt securities Government grants Venture capital M&A are all ways to _______
raise capital
Classification or regression model that improves the accuracy of a simple decision tree by generating multiple decision trees and taking a majority vote of them to predict the output, which is a continuous variable (eg, age) for a regression problem and a discrete variable (eg, either black, white, or red) for classification is called ___ ___ and is a type of ______ learning
random forest, supervised
the development of medications based on the study of the structures and functions of target molecules. It is a process in which finding of new medication based on knowledge of biological target is done. It involves design of small molecules that are complementary in shape and charge to biomolecular target
rational drug design
Often uses cluster behavior prediction to identify the important data necessary for making a recommendation is called ___ ___ and is a type of ______ learning
recommender system, unsupervised
two main uses of blockchain:
record-keeping, transacting
A multilayered neural network that can store information in context nodes, allowing it to learn data sequences and output a number or another sequence
recurrent neural network
if you are working with time-series data or sequences (eg. audio recordings or text) you would use
recurrent neural network
Blockchains short-term impact will be ____ ____ long term impact: transforming _____ _______
reducing, business models
Optimize the trading strategy for an options-trading portfolio Balance the load of electricity grids in varying demand cycles Stock and pick inventory using robots Optimize the driving behavior of self-driving cars Optimize pricing in real time for an online auction of a product with limited supply are all examples of _______ ________
reinforcement learning
An algorithm learns to perform a task simply by trying to maximize rewards it receives for its actions (eg, maximizes points it receives for increasing returns of an investment portfolio) do not have an ideal end state; only way to figure out what is going on with environment is to engage with it (online chess)
reinforcement learning
Incentives to improve adherence: These trials suggested that financial incentives have the potential to improve adherence. The ____ and _______ of this approach to improve adherence is a barrier, and most clinicians and health systems will not have the resources to provide significant incentives to improve adherence.
scalability, feasibility
The MediLedger Project, backed by a group of companies that includes pharma giants Genentech and Pfizer, has successfully piloted a program that uses blockchain tools provided by Chronicled, a US-based startup, to track medicines. is an example of _____ _____ ______ using blockchain technology
securing supply chains
CBT: Clinicians could consider referring ______ patients with nonadherence for cognitive behavioral therapy, if cognitive behavioral therapy was available within their practice or health system. However, these trials suggested that adherence is _______ to improve unless such interventions are delivered by trained specialists over at least ______ sessions, raising concerns about ____ and ________.
selected, unlikely, 4, feasibility, scalability
Model in which artificial neurons (software- based calculators) make up three layers (an input layer, a hidden layer where calculations take place, and an output layer) that can be used to classify data or find the relationship between variables in regression problems is called ___ ___ and is a type of ______ learning
simple neural network, supervised
While the future of personalized medicine belongs to big data from genomics and even proteomics, the present revenue realities in genetic diagnostics are still focused on _____ gene panels.
small
medication-taking regimens: Improvements were ______ and may not translate into clinically important effects. Refill reminders may be most effective when they are targeted to patients who do not fill a new prescription at the time of ____ ________. Electronic drug monitors with ______ reminders may also be effective, although many practices and health systems may not have the ______ or resources available to provide them to nonadherent patients.
small, disease diagnosis, automated, infrastructure
Which quantum computing technology are big businesses investing in most?
superconducting qubits
quantum computing technology: investors from google, IBM, etc. The basic element is a two-level energy system of a superconducting circuit which forms a somewhat noise-resistant qubit
superconducting qubits
Two of the leading quantum-computing technologies: ____________ ____________
superconducting qubits, ion traps
linear regression, Logistic regression, linear/ quadratic discriminant analysis, decision tree, naive bayes, support vector machines, random forest, Ada boost, gradient-boosting tree, simple neural network are all types of _____ learning
supervised
An algorithm uses training data and feedback from humans to learn the relationship of given inputs to a given output (eg, how the inputs "time of year" and "interest rates" predict housing prices) you know how to classify data
supervised learning
three types of machine learning:
supervised, unsupervised, reinforcement
4 promising healthcare related scenarios for blockchain: securing _____ ______ putting patients in charge of their ______ accelerating _____ transforming _____ _______
supply chains, data, R&D, service delivery
A technique that's typically used for classification but can be transformed to perform regression. It draws an optimal division between classes (as wide as possible). It also can be quickly generalized to solve nonlinear problems is called ___ ___ and is a type of ______ learning
support vector machine, supervised
is based on the understanding that the whole is greater than the sum of the parts. It has been responsible for some of the most important developments in the science of human health and environmental sustainability. It is a holistic approach to deciphering the complexity of biological system that starts from the understanding that the networks that form the whole of living organisms are more than the sum of their parts
system biology
problems with ____ and ______ in blockchain: Perhaps most significant, in order to mature beyond proof of concept and deliver on their promise, actual applications must be built on scalable protocols. Finding the skills to build the right technology is a major challenge today. Competition for talent among companies working in multiple ad- vanced technologies (such as artificial in- telligence and machine learning) means that the supply of skills and expertise will remain a bottleneck for the foreseeable future
technology, talent
health apps and ______ claim to address oncology-related _____ communication chains these solutions offer more potential when partnered with care teams and _____ EMRS
telemedicine, fragmented, integrated
Blockchain can enable entirely new delivery models that speed up services, reduce costs, and potentially disrupt long-standing procedures. One possibility is a prescription drug exchange—a patient-driven prescription fulfillment system independent of any one pharmacy or pharmacy benefit manager. Such a model would lower the cost of prescriptions by opening market competition to small and new large players, removing barriers to entry, and streamlining how companies comply with formula regulation. (no concrete solution yet) is an example of _____ ______ ______ using blockchain technology
transforming service delivery
An algorithm explores input data without being given an explicit output variable (eg, explores customer demographic data to identify patterns) you don't know how to classify data
unsupervised learning
_____ era + ______ health era= THE FUTURE OF PATIENT CARE IN __________
value, digital, oncology
blockchain: private/ permissioned=
very high scalability
Incorporating adherence questions into clinic _______ or rooming procedures by medical assistants are feasible and effective means of identifying patients who may be non- adherent. Clinicians should also be aware of patient-specific factors, such as socioeconomic status, that may affect the choice of intervention. These trade-offs will require a conversation to elicit a patient's barriers to adherence and use of shared decision-making to choose an intervention that is acceptable.
waiting rooms,