The rise of artificial intelligence in healthcare applications

In the paper

“However, mining of the large-scale chemistry data is needed to efficiently classify potential drug compounds and machine learning techniques have shown great potential [15].”

What the sources say

• Supported

  In “The tipping point for machine learning,” the paper describes large chemical and biological datasets as valuable for drug-discovery machine learning and says they must be prepared for modeling. “Machine learning models in action” describes using these methods to virtually screen and select compounds, supporting the citing sentence’s broad paraphrase. (LLM, conf=0.88)

  cites [15] Ekins, 2019, Exploiting machine learning for end-to-end drug discovery and development, Nat Mater source

In the paper

“An example of this is a group at the Institute of Information Technology at the Alpen-Adria Universität Klagenfurt that uses surgery videos as training material in order to identify a specific intervention made by the surgeon. For example, when an act of dissection or cutting is performed on the patient’s tissues or organs, the algorithm recognizes the likelihood of the intervention as well as the”

What the sources say

• Supported

  The Introduction describes training classifiers on annotated gynecological surgery videos to predict surgical-action and anatomical-structure classes. The Action section includes blunt dissection and cutting among the actions classified. The paper supports identifying an anatomical structure in a video scene, rather than pinpointing its location within the frame. (LLM, conf=0.90)

  cites [27] Petscharnig, 2018, Learning laparoscopic video shot classification for gynecological surgery, Multimed Tools Appl source

In the paper

“These elements are used to motivate and drive users to reach their goals [34].”

What the sources say

• Supported

  The Introduction (‘Gamification: a new model?’) describes game-design elements as motivational features, including goal-setting and progress feedback. It also says gamification aims to drive behaviour through the motivating qualities of games, supporting the sentence as a statement of purpose. (LLM, conf=0.91)

  cites [34] Johnson, 2016, Gamification for health and wellbeing: a systematic review of the literature, Internet Interv source

In the paper

“Current AI applications within healthcare are often standalone applications, these are often used for diagnostics using medical imaging and for disease prediction using remote patient monitoring [38].”

What the sources say

• Partly supported

  The Introduction (page not provided) says remote monitoring and trend analysis can detect deterioration in a patient’s condition early. It does not describe AI-based disease prediction, standalone AI applications, or diagnosis using medical imaging; its mention of transmitted still images is not evidence of AI imaging diagnostics. (LLM, conf=0.82)

  cites [38] Rojahn, 2016, Remote monitoring of chronic diseases: a landscape assessment of policies in four European countries, PLoS One sourceExperimental — accuracy not yet measured

In the paper

“Another platform called MoleculeNet is also used to translate two-dimensional molecular structures into novel features/descriptors, which can then be used in predicting toxicity of the given molecule. The MoleculeNet platform is built on data from various public databases and more than 700,000 compounds have already been tested for toxicity or other properties [20].”

What the sources say

• Supported

  Split into 2 atoms — 2 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  cites [20] Wu, 2018, MoleculeNet: a benchmark for molecular machine learning, Chem Sci source

In the paper

“Merk et al. trained a recurrent neural network to capture a large number of bioactive compounds such as SMILES strings. This model was then fine-tuned to recognize retinoid X and peroxisome proliferator-activated receptor agonists. The identified compounds were synthesized and demonstrated potent receptor modulatory activity in in vitro assays [23].”

What the sources say

• Supported

  The paper’s Abstract states that the authors trained an RNN on bioactive compounds represented as SMILES strings and fine-tuned it for RXR and PPAR agonists. It reports that five model-designed compounds were synthesized and four showed nanomolar to low-micromolar receptor-modulatory activity in cell-based assays; the claim should be read with that four-of-five qualification. (LLM, conf=0.91)

  cites [23] Merk, 2018, De novo design of bioactive small molecules by artificial intelligence, Mol Inform source

In the paper

“Machine learning has also been implemented to assess the toxicity of molecules, for instance, using DeepTox, a DL-based model for evaluating the toxic effects of compounds based on a dataset containing many drug molecules [19].”

What the sources say

• Supported

  The Abstract describes DeepTox as a deep-learning pipeline for toxicity prediction and says the Tox21 challenge comprised 12,000 environmental chemicals and drugs tested for toxic effects. Section 2.3 describes how the pipeline uses compound data to generate predictions. (LLM, conf=0.91)

  cites [19] Mayr, 2016, DeepTox: toxicity prediction using deep learning, Front Environ Sci source

In the paper

“Studying data from a cohort of diabetic and mental health patients it was demonstrated that DeepCare could predict the progression of disease, optimal interventions, and assessing the likelihood for readmission [37].”

What the sources say

• Partly supported

  The Conclusion reports that DeepCare was demonstrated on predicting next disease stages, recommending interventions, and estimating unplanned readmission among diabetic and mental health patients. But §4.3 evaluates prediction of interventions recorded for current diagnoses; it does not establish that the recommended interventions were optimal. (LLM, conf=0.89)

  cites [37] Pham, 2017, Predicting healthcare trajectories from medical records: a deep learning approach, J Biomed Inform sourceExperimental — accuracy not yet measured

In the paper

“One main pain point refers to the lack of empathy and the ethical dilemma that can occur between AI and patients [70].”

What the sources say

• Supported

  In Results, “Empathy and Communication” identifies inability to provide empathic care as a main perceived limitation of future technology. Results, “Social and Ethical Concerns,” also records concerns about safety, responsibility, and liability; the Discussion identifies risks including bias and patient privacy. These are concerns reported by GPs, not demonstrated harms in AI–patient interactions. (LLM, conf=0.84)

  cites [70] Blease, 2019, Artificial intelligence and the future of primary care: exploratory qualitative study of UK general practitioners’ views, J Med Internet Res source

In the paper

“In a recent study, when asked about the future of AI on primary care, while acknowledging its potential benefits, most practitioners were extremely skeptical regarding it playing a significant role in the future of the profession.”

What the sources say

• Supported

  The Discussion’s Principal Findings say that skepticism was the dominant view among respondents and that most GPs expected technological innovation to have a restricted scope in general practice. The Results’ Benefits of Future Technology in Primary Care section also reports perceived benefits, chiefly improved efficiency and reduced administrative work. (LLM, conf=0.92)

  cites [70] Blease, 2019, Artificial intelligence and the future of primary care: exploratory qualitative study of UK general practitioners’ views, J Med Internet Res source

In the paper

“The AI-associated healthcare market is expected to grow rapidly and reach USD 6.6 billion by 2021 corresponding to a 40% compound annual growth rate [4].”

What the sources say

• Not checked yetcites [4] Bresnick J. Artificial intelligence in healthcare market to see 40% CAGR surge; 2017.

In the paper

“Healthcare institutions are also fighting to keep up with all the new technological developments and the high expectations of patients with respect to levels of service and outcomes as they know it from consumer products including those of Amazon and Apple [2].”

What the sources say

• ?Not checked — no full text of the sourcecites [2] Kirch, 2017, Addressing the physician shortage: the peril of ignoring demography, JAMA source

In the paper

“For these patients, this immersive experience could act as a personal rehabilitation physiotherapist who engages their upper limb movement multiple times a day, allowing for possible neuroplasticity and a gradual return of normal motor function to these regions.”

What the sources say

• ?Not checked — no full text of the sourcecites [31] Lee, 2020, Upper extremity rehabilitation using fully immersive virtual reality games with a head mount display: a feasibility study, PM R source

In the paper

“There are many areas in healthcare in which NLP can provide substantial benefits. Some of the more immediate applications include [36]”

What the sources say

• Not checked yetcites [36] Rangasamy ASS, Nadenichek R, Rayasam M. Natural language processing in healthcare; 2018.

In the paper

“As an example, the program druGAN used adversarial autoencoders to generate new molecular fingerprints and drug designs incorporating features such as solubility and absorption based on predefined anticancer drug properties. These results suggest a substantial improvement in the efficiency in generating new drug designs with specific properties [21].”

What the sources say

• ?Not checked — no full text of the sourcecites [21] Kadurin, 2017, druGAN: an advanced generative adversarial autoencoder model for de novo generation of new molecules with desired molecular properties in silico, Mol Pharm source

In the paper

“Advantages of precision medicine include reduced healthcare costs, reduction in adverse drug response, and enhancing effectivity of drug action [11].”

What the sources say

• ?Not checked — no full text of the sourcecites [11] Konieczny L, Roterman I. Personalized precision medicine. Bio-Algorithms Med-Syst 2019; 15. source

In the paper

“In this way, molecular properties including octanol, solubility melting point, and biological activity can be evaluated as demonstrated by Coley et al. and others and be used to predict new features of the drug molecules [18].”

What the sources say

• ?Not checked — no full text of the sourcecites [18] Coley, 2017, Convolutional embedding of attributed molecular graphs for physical property prediction, J Chem Inf Model source

In the paper

“So game-like technologies such as VR and AR could enhance and enrich the learning experience for future medical and health-related disciplines [29].”

What the sources say

• Not checked yetcites [29] Chien CH, Chen CH, Jeng TS. An interactive augmented reality system for learning anatomy structure. In: Proceedings of the International MultiConference of Engineers and Computer Scientists 2010, IMECS 2010; 2010. http://www.iaeng.org/publication/IMECS2018/

In the paper

“A study has shown that late-stage adult cancer patients can use this technology with minimum physical discomfort and in return benefit from an enhanced relaxed state, entertainment, and a much-needed distraction [32].”

What the sources say

• Not checked yetcites [32] Baños, 2013, A positive psychological intervention using virtual reality for patients with advanced cancer in a hospital setting: a pilot study to assess feasibility, Support Care Cancer

In the paper

“The drug molecules and the associated features used in the in silico models are transformed into vector format so they can be read by the learning systems. Generally, the data used here include molecular descriptors (e.g., physicochemical properties) and molecular fingerprints (molecular structure) as well as simplified molecular input line entry system (SMILES) strings and grids for convolutional”

What the sources say

• ?Not checked — no full text of the sourcecites [17] Lavecchia, 2019, Deep learning in drug discovery: opportunities, challenges and future prospects, Drug Discov Today source

In the paper

“Remote monitoring and picking up on early signs of disease could be immensely beneficial for those who suffer from chronic conditions and the elderly.”

In the paper

“For real-life and precise execution, AR would be more advantageous in healthcare settings.”

In the paper

“According to the researchers, the “model sees sequences of words/speaking style” and decides whether these emerging patterns are likely to be seen in individuals with and without depression [63].”

What the sources say

• Not checked yetcites [63] Matheson R. Model can more naturally detect depression in conversations. MIT News; 2018.

In the paper

“A review by Becker [62] suggests that AI used in healthcare can serve clinicians, patients, and other healthcare workers in four different ways. Here, we will use these suggestions as inspirations and will expand on their contribution toward a successful implementation of AI in healthcare: (Fig. 2.9 )1.Assessment of disease onset and treatment success.2.Management or alleviation of complications.3”

What the sources say

• ?Not checked — no full text of the sourcecites [62] Becker, 2019, Artificial intelligence in medicine: what is it doing for us today?, Health Policy Technol source

In the paper

“Machine learning algorithms used in smart home applications include probabilistic and discriminative methods such as Naive Bayes classifier and Hidden Markov Model, support vector machine, and artificial neural networks [54].”

What the sources say

• ?Not checked — no full text of the sourcecites [54] Fahad LG, Ali A, Rajarajan M. Learning models for activity recognition in smart homes. In: Information science and applications. Berlin: Springer; 2015. p. 819–26. source

In the paper

“The tactile sensory system can detect mass calcifications inside the breast tissue based on palpation of different points of the tissue and comparing with different reference data, and subsequently determine whether there are any significant abnormalities in the breast tissue.”

What the sources say

• ?Not checked — no full text of the sourcecites [44] Naidu, 2017, A breakthrough in tumor localization: combining tactile sensing and ultrasound to improve tumor localization in robotics-assisted minimally invasive surgery, IEEE Robot Autom Mag source

In the paper

“The Kompaï robot used for the MARIO project was developed by Robosoft and is a robot containing a camera, a Kinect motion sensor, and two LiDAR remote sensing systems for navigation and object identification [58].”

What the sources say

• ?Not checked — no full text of the sourcecites [58] D’Onofrio, 2019, MARIO Project: validation and evidence of service robots for older people with dementia, J Alzheimers Dis source

In the paper

“Assistive technologies increase the self-dependencies of patients, encouraging user participation in Information and Communication Technology (ICT) tools to provide remote care services type assistance and provide information to the healthcare professionals.”

What the sources say

• ?Not checked — no full text of the sourcecites [51] Anderson, 2015, The impact of assistive technologies on formal and informal home care, Gerontologist source

In the paper

“Altogether, studies investigating cognitive stimulation seem to demonstrate a decrease in the rate of cognitive decline and progression of dementia.”

In the paper

“In a study published in the journal of trauma and acute care surgery, researchers think that by understanding the microbiological niches (biomarkers) of trauma patients, we could hold the key to future wound infections and therefore can allow healthcare workers to take the necessary arrangements to prevent the worst outcome [64].”

What the sources say

• ?Not checked — no full text of the sourcecites [64] Dente, 2017, Towards precision medicine: accurate predictive modeling of infectious complications in combat casualties, J Trauma Acute Care Surg source

In the paper

“Virtrael is one of such cognitive stimulation platforms and serves to assess, stimulate, and train various cognitive skills that experience a decline in the patient. The Virtrael program is based on visual memory training and the project is carried out by three different key functionalities: configuration, communication, and games.”

What the sources say

• ?Not checked — no full text of the sourcecites [60] Garcia-Alonso J, Fonseca C, editors. Gerontechnology: First International Workshop. In: First international workshop on gerotechnology. Springer; 2018. source

In the paper

“There are various international efforts in gathering EMR data across countries including Observational Health Data Science and Informatics (OHDSI), who have consolidated 1.26 billion patient records from 17 different countries [40].”

What the sources say

• ?Not checked — no full text of the sourcecites [40] Wang, 2019, Deep learning in medicine—promise, progress, and challenges, JAMA Intern Med source

In the paper

“AAL applications typically collect data through sensors and cameras and apply various artificially intelligent tools for developing an intelligent system [52].”

What the sources say

• ?Not checked — no full text of the sourcecites [52] Barnay, 2016, Does home care for dependent elderly people improve their mental health?, J Health Econ source

In the paper

“Most people want to preserve their autonomy, even at an old age, and maintain control over their lives and decisions [50].”

What the sources say

• Not checked yetcites [50] Roberts AW, Ogunwole SU, Blakeslee L, Rabe MA, The population 65 years and older in the United States: 2016; 2018.

In the paper

“The games are intended to train cognitive skills of the patient including memory, attention, and planning (Fig. 2.8 ) [60].”

What the sources say

• ?Not checked — no full text of the sourcecites [60] Garcia-Alonso J, Fonseca C, editors. Gerontechnology: First International Workshop. In: First international workshop on gerotechnology. Springer; 2018. source

In the paper

“Using the stored data, the framework of DeepCare can model disease progression, support intervention recommendation, and provide disease prognosis based on EMR databases.”

What the sources say

• ?Not checked — no full text of the sourcecites [41] Pham T, Tran T, Phung D, Venkatesh S. DeepCare: a deep dynamic memory model for predictive medicine. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer; 2016. source

In the paper

“By daily collection of patient data, activities of daily living are defined over time and abnormalities can be detected as a deviation from the routine.”

In the paper

“However, integrating such standalone applications with EMR data could provide even greater value by adding personal medical data and history as well as a large statistical reference library to make classifications and predictions more accurate and powerful.”

In the paper

“The robotic pet PARO, a baby seal robot, is the most widely used robotic pet and carries various sensors to sense touch, sounds, and visual objects [61].”

What the sources say

• ?Not checked — no full text of the sourcecites [61] Vitanza A, D’Onofrio G, Ricciardi F, Sancarlo D, Greco A, Giuliani F. Assistive robots for the elderly: innovative tools to gather health relevant data. In: Data science for healthcare: methodologies and applications. Springer; 2019. source

In the paper

“Machine learning techniques can also contribute toward the prediction of serious complications such as neuropathy that could arise for those suffering from type 2 diabetes or early cardiovascular irregularities.”

In the paper

“Patient-care assistance technologies can improve the workflow for clinicians and contribute toward patient’s autonomy and well-being.”

In the paper

“It is also very important to note that at the time of an epidemic, an outbreak, natural or manmade disaster, or simply when the patient is away from their usual dwelling, a technology that allows humans to remotely interact and solve problems will have to become a necessity.”

In the paper

“Speech and text recognition are already employed for tasks like patient communication and capture of clinical notes, and their usage will increase.”

In the paper

“It also seems increasingly clear that AI systems will not replace human clinicians on a large scale, but rather will augment their efforts to care for patients.”

In the paper

“Perhaps the only healthcare providers who will risk their careers over time may be those who refuse to work alongside AI.”

In the paper

“For widespread adoption to take place, AI systems must be approved by regulators, integrated with EHR systems, standardized to a sufficient degree that similar products work in a similar fashion, taught to clinicians, paid for by public or private payer organizations, and updated over time in the field.”

In the paper

“These challenges will ultimately be overcome, but they will take much longer to do so than it will take for the technologies themselves to mature.”

In the paper

“If each patient is treated as an independent system, then based on the variety of designated data available, a bespoke approach can be implemented.”

In the paper

“Over time, human clinicians may move toward tasks and job designs that draw on uniquely human skills like empathy, persuasion, and big-picture integration.”

In the paper

“This setting can be applied both at home and in a hospital setting to relieve work pressure from healthcare workers and improve service.”

In the paper

“As a result, we expect to see limited use of AI in clinical practice within 5 years and more extensive use within 10 years.”

In the paper

“Combine this with the ever-increasing breakthroughs in the field of smart healthcare materials [71] and AI, one could envisage patients managing most of their own conditions at home and when necessary get in touch with a relevant healthcare worker who will refer them to more specialized physicians who could tend to their needs.”

What the sources say

• ?Not checked — no full text of the sourcecites [71] Zang, 2015, Advances of flexible pressure sensors toward artificial intelligence and health care applications, Mater Horiz source

In the paper

“Here, patients will be able to interact with the device in a remote manner and access their biometric data, all the while feeling that they are interacting with a caring and empathetic system that truly wants the best outcome for them.”

In the paper

“Although early efforts at providing diagnosis and treatment recommendations have proven challenging, we expect that AI will ultimately master that domain as well.”

In the paper

“AI can accelerate the diagnosis process and medical research.”