Current status of artificial intelligence methods for skin cancer survival analysis: a scoping review.

Skin cancer mortality rates continue to rise, and survival analysis is increasingly needed to understand who is at risk and what interventions improve outcomes. However, current statistical methods are limited by inability to synthesize multiple data types, such as patient genetics, clinical history, demographics, and pathology and reveal significant multimodal relationships through predictive algorithms. Advances in computing power and data science enabled the rise of artificial intelligence (AI), which synthesizes vast amounts of data and applies algorithms that enable personalized diagnostic approaches. Here, we analyze AI methods used in skin cancer survival analysis, focusing on supervised learning, unsupervised learning, deep learning, and natural language processing. We illustrate strengths and weaknesses of these approaches with examples. Our PubMed search yielded 14 publications meeting inclusion criteria for this scoping review. Most publications focused on melanoma, particularly histopathologic interpretation with deep learning. Such concentration on a single type of skin cancer amid increasing focus on deep learning highlight growing areas for innovation; however, it also demonstrates opportunity for additional analysis that addresses other types of cutaneous malignancies and expands the scope of prognostication to combine both genetic, histopathologic, and clinical data. Moreover, researchers may leverage multiple AI methods for enhanced benefit in analyses. Expanding AI to this arena may enable improved survival analysis, targeted treatments, and outcomes.

Prevalence of potentially harmful multidrug interactions on medication lists of elderly ambulatory patients.

BACKGROUND: It has been hypothesized that polypharmacy may increase the frequency of multidrug interactions (MDIs) where one drug interacts with two or more other drugs, amplifying the risk of associated adverse drug events (ADEs). The main objective of this study was to determine the prevalence of MDIs in medication lists of elderly ambulatory patients and to identify the medications most commonly involved in MDIs that amplify the risk of ADEs. METHODS: Medication lists stored in the electronic health record (EHR) of 6,545 outpatients ≥60 years old were extracted from the enterprise data warehouse. Network analysis identified patients with three or more interacting medications from their medication lists. Potentially harmful interactions were identified from the enterprise drug-drug interaction alerting system. MDIs were considered to amplify the risk if interactions could increase the probability of ADEs. RESULTS: MDIs were identified in 1.3 % of the medication lists, the majority of which involved three interacting drugs (75.6 %) while the remainder involved four (15.6 %) or five or more (8.9 %) interacting drugs. The average number of medications on the lists was 3.1 ± 2.3 in patients with no drug interactions and 8.6 ± 3.4 in patients with MDIs. The prevalence of MDIs on medication lists was greater than 10 % in patients prescribed bupropion, tramadol, trazodone, cyclobenzaprine, fluoxetine, ondansetron, or quetiapine and greater than 20 % in patients prescribed amiodarone or methotrexate. All MDIs were potentially risk-amplifying due to pharmacodynamic interactions, where three or more medications were associated with the same ADE, or pharmacokinetic, where two or more drugs reduced the metabolism of a third drug. The most common drugs involved in MDIs were psychotropic, comprising 35.1 % of all drugs involved. The most common serious potential ADEs associated with the interactions were serotonin syndrome, seizures, prolonged QT interval and bleeding. CONCLUSIONS: An identifiable number of medications, the majority of which are psychotropic, may be involved in MDIs in elderly ambulatory patients which may amplify the risk of serious ADEs. To mitigate the risk, providers will need to pay special attention to the overlapping drug-drug interactions which result in MDIs.

Clinical Features and Outcomes of Patients with Dementia Compared to an Aging Cohort Hospitalized During the Initial New York City COVID-19 Wave.

BACKGROUND: Patients with dementia are vulnerable during the coronavirus disease 2019 (COVID-19) pandemic, yet few studies describe their hospital course and outcomes. OBJECTIVE: To describe and compare the hospital course for COVID-19 patients with dementia to an aging cohort without dementia in a large New York City academic medical center. METHODS: This was a single-center retrospective cohort study describing all consecutive patients age 65 or older with confirmed COVID-19 who presented to the emergency department or were hospitalized at New York-Presbyterian/Columbia University Irving Medical Center between March 6 and April 7, 2020. RESULTS: A total of 531 patients were evaluated, including 116 (21.8%) with previously diagnosed dementia, and 415 without dementia. Patients with dementia had higher mortality (50.0%versus 35.4%, p = 0.006); despite similar comorbidities and complications, multivariate analysis indicated the association was dependent on age, sex, comorbidities, and code status. Patients with dementia more often presented with delirium (36.2%versus 11.6%, p < 0.001) but less often presented with multiple other COVID-19 symptoms, and these findings remained after adjusting for age and sex. CONCLUSION: Hospitalized COVID-19 patients with dementia had higher mortality, but dementia was not an independent risk factor for death. These patients were approximately 3 times more likely to present with delirium but less often manifested or communicated other common COVID-19 symptoms. For this high-risk population in a worsening pandemic, understanding the unique manifestations and course in dementia and aging populations may help guide earlier diagnosis and optimize medical management.

Management of afebrile neonates with pustules and vesicles in a pediatric emergency department.

OBJECTIVES: To assess the management and outcomes of vesicles and pustules in afebrile neonates presenting to the pediatric emergency department. METHODS: Using International Classification of Diseases, Ninth Revision, codes, we identified patients 0-60 days old presenting to our pediatric emergency department from 2008 to 2015 with a possible diagnosis of pustules or vesicles. We then used natural language processing followed by manual chart review to identify afebrile neonates with pustules or vesicles. We collected clinical data from the electronic medical record. We also assessed current practice patterns for neonatal pustules or vesicles using a survey administered to attending physicians. RESULTS: Of the 971 possible cases identified using International Classification of Diseases, Ninth Revision, codes for fluid-filled lesions, only 64 patients had vesicles (n = 9) and pustules (n = 55). One-third (22/64) of afebrile neonates with pustules and vesicles were admitted to the hospital and received empiric parenteral therapy. Admission, parenteral antibiotics, and antiviral therapy were more common in neonates presenting with vesicles than in those with pustules alone. Apart from 2 presumed blood culture contaminants, there were no positive blood or cerebrospinal fluid cultures. Two patients had positive urine cultures. Institutional survey data showed practice patterns consistent with these retrospective results. CONCLUSION: Although one-third of neonates with pustules and vesicles were admitted to the hospital and received parenteral therapy, there were no cerebrospinal fluid or blood infections or any confirmed evidence of herpes simplex virus disease. These findings suggest that afebrile, well-appearing neonates presenting with pustules alone may not need a full serious bacterial infection examination. Larger studies are needed to confirm these findings and assess outcomes, especially in afebrile neonates with vesicles.

Monitoring prescribing patterns using regression and electronic health records.

BACKGROUND: It is beneficial for health care institutions to monitor physician prescribing patterns to ensure that high-quality and cost-effective care is being provided to patients. However, detecting treatment patterns within an institution is challenging, given that medications and conditions are often not explicitly linked in the health record. Here we demonstrate the use of statistical methods together with data from the electronic health care record (EHR) to analyze prescribing patterns at an institution. METHODS: As a demonstration of our method, which is based on regression, we collect EHR data from outpatient notes and use a case/control study design to determine the medications that are associated with hypertension. We also use regression to determine which conditions are associated with a preferential use of one or more classes of hypertension agents. Finally, we compare our method to methods based on tabulation. RESULTS: Our results show that regression methods provide more reasonable and useful results than tabulation, and successfully distinguish between medications that treat hypertension and medications that do not. These methods also provide insight into in which circumstances certain drugs are preferred over others. CONCLUSIONS: Our method can be used by health care institutions to monitor physician prescribing patterns and ensure the appropriateness of treatment.

Early recognition of multiple sclerosis using natural language processing of the electronic health record.

BACKGROUND: Diagnostic accuracy might be improved by algorithms that searched patients' clinical notes in the electronic health record (EHR) for signs and symptoms of diseases such as multiple sclerosis (MS). The focus this study was to determine if patients with MS could be identified from their clinical notes prior to the initial recognition by their healthcare providers. METHODS: An MS-enriched cohort of patients with well-established MS (n = 165) and controls (n = 545), was generated from the adult outpatient clinic. A random sample cohort was generated from randomly selected patients (n = 2289) from the same adult outpatient clinic, some of whom had MS (n = 16). Patients' notes were extracted from the data warehouse and signs and symptoms mapped to UMLS terms using MedLEE. Approximately 1000 MS-related terms occurred significantly more frequently in MS patients' notes than controls'. Synonymous terms were manually clustered into 50 buckets and used as classification features. Patients were classified as MS or not using Naïve Bayes classification. RESULTS: Classification of patients known to have MS using notes of the MS-enriched cohort entered after the initial ICD9[MS] code yielded an ROC AUC, sensitivity, and specificity of 0.90 [0.87-0.93], 0.75[0.66-0.82], and 0.91 [0.87-0.93], respectively. Similar classification accuracy was achieved using the notes from the random sample cohort. Classification of patients not yet known to have MS using notes of the MS-enriched cohort entered before the initial ICD9[MS] documentation identified 40% [23-59%] as having MS. Manual review of the EHR of 45 patients of the random sample cohort classified as having MS but lacking an ICD9[MS] code identified four who might have unrecognized MS. CONCLUSIONS: Diagnostic accuracy might be improved by mining patients' clinical notes for signs and symptoms of specific diseases using NLP. Using this approach, we identified patients with MS early in the course of their disease which could potentially shorten the time to diagnosis. This approach could also be applied to other diseases often missed by primary care providers such as cancer. Whether implementing computerized diagnostic support ultimately shortens the time from earliest symptoms to formal recognition of the disease remains to be seen.

Medication-indication knowledge bases: a systematic review and critical appraisal.

OBJECTIVE: Medication-indication information is a key part of the information needed for providing decision support for and promoting appropriate use of medications. However, this information is not readily available to end users, and a lot of the resources only contain this information in unstructured form (free text). A number of public knowledge bases (KBs) containing structured medication-indication information have been developed over the years, but a direct comparison of these resources has not yet been conducted. MATERIAL AND METHODS: We conducted a systematic review of the literature to identify all medication-indication KBs and critically appraised these resources in terms of their scope as well as their support for complex indication information. RESULTS: We identified 7 KBs containing medication-indication data. They notably differed from each other in terms of their scope, coverage for on- or off-label indications, source of information, and choice of terminologies for representing the knowledge. The majority of KBs had issues with granularity of the indications as well as with representing duration of therapy, primary choice of treatment, and comedications or comorbidities. DISCUSSION AND CONCLUSION: This is the first study directly comparing public KBs of medication indications. We identified several gaps in the existing resources, which can motivate future research.

Presence of early CKD-related metabolic complications predict progression of stage 3 CKD: a case-controlled study.

BACKGROUND: Only a subset of patients who enter stage 3 chronic kidney disease (CKD) progress to stage 4. Identifying which patients entering stage 3 are most likely to progress could improve outcomes, by allowing more appropriate referrals for specialist care, and spare those unlikely to progress the adverse effects and costliness of an unnecessarily aggressive approach. We hypothesized that compared to non-progressors, patients who enter stage 3 CKD and ultimately progress have experienced greater loss of renal function, manifested by impairment of metabolic function (anemia, worsening acidosis and mineral abnormalities), than is reflected in the eGFR at entry to stage 3. The purpose of this case-controlled study was to design a prediction model for CKD progression using laboratory values reflecting metabolic status. METHODS: Using data extracted from the electronic health record (EHR), two cohorts of patients in stage 3 were identified: progressors (eGFR declined >3 ml/min/1.73 m2/year; n=117) and non-progressors (eGFR declined <1 ml/min/1.713 m2; n=364). Initial laboratory values recorded a year before to a year after the time of entry to stage 3, reflecting metabolic complications (hemoglobin, bicarbonate, calcium, phosphorous, and albumin) were obtained. Average values in progressors and non-progressors were compared. Classification algorithms (Naïve Bayes and Logistic Regression) were used to develop prediction models of progression based on the initial lab data. RESULTS: At the entry to stage 3 CKD, hemoglobin, bicarbonate, calcium, and albumin values were significantly lower and phosphate values significantly higher in progressors compared to non-progressors even though initial eGFR values were similar. The differences were sufficiently large that a prediction model of progression could be developed based on these values. Post-test probability of progression in patients classified as progressors or non-progressors were 81% (73% - 86%) and 17% (13% - 23%), respectively. CONCLUSIONS: Our studies demonstrate that patients who enter stage 3 and ultimately progress to stage 4 manifest a greater degree of metabolic complications than those who remain stable at the onset of stage 3 when eGFR values are equivalent. Lab values (hemoglobin, bicarbonate, phosphorous, calcium and albumin) are sufficiently different between the two cohorts that a reasonably accurate predictive model can be developed.

Vitamin D resistance in chronic kidney disease (CKD).

BACKGROUND: Previous studies have shown that treatment with ergocalciferol in patients with CKD stage 3 + 4 is not effective with less than 33% of patients achieving a 25-OH vitamin D target of >30 ng/ml. The aim of this study was to test the response to cholecalciferol in CKD. We attempted to replete 25-OH vitamin D to a target level of 40-60 ng/ml using the response to treatment and PTH suppression as an outcome measure. METHODS: This retrospective cohort study identified patients (Stages 2-5 and Transplant) from 2001-2010 who registered at the Chronic Kidney Disease Clinic. Patients received cholecalciferol 10,000 IU capsules weekly as initial therapy. When levels above 40 ng/ml were not achieved, doses were titrated up to a maximum of 50,000 IU weekly. Active vitamin D analogs were also used in some Stage 4-5 CKD patients per practice guidelines. Patients reaching at least one level of 40 ng/mL were designated RESPONDER, and if no level above 40 ng/mL they were designated NON-RESPONDER. Patients were followed for at least 6 months and up to 5 years. RESULTS: 352 patients were included with a mean follow up of 2.4 years. Of the CKD patients, the initial 25-OH vitamin D in the NON-RESPONDER group was lower than the RESPONDER group (18 vs. 23 ng/ml) (p = 0.03). Among all patients, the initial eGFR in the RESPONDER group was significantly higher than the NON-RESPONDER group (36 vs. 30 ml/min/1.73 m2) (p < 0.001). Over time, the eGFR of the RESPONDER group stabilized or increased (p < 0.001). Over time, the eGFR in the NON-RESPONDER group decreased toward a trajectory of ESRD. Proteinuria did not impact the response to 25-OH vitamin D replacement therapy. There were no identifiable variables associated with the response or lack of response to cholecalciferol treatment. CONCLUSIONS: CKD patients treated with cholecalciferol experience treatment resistance in raising vitamin D levels to a pre-selected target level. The mechanism of vitamin D resistance remains unknown and is associated with progressive loss of eGFR. Proteinuria modifies but does not account for the vitamin D resistance.

Combing signals from spontaneous reports and electronic health records for detection of adverse drug reactions.

OBJECTIVE: Data-mining algorithms that can produce accurate signals of potentially novel adverse drug reactions (ADRs) are a central component of pharmacovigilance. We propose a signal-detection strategy that combines the adverse event reporting system (AERS) of the Food and Drug Administration and electronic health records (EHRs) by requiring signaling in both sources. We claim that this approach leads to improved accuracy of signal detection when the goal is to produce a highly selective ranked set of candidate ADRs. MATERIALS AND METHODS: Our investigation was based on over 4 million AERS reports and information extracted from 1.2 million EHR narratives. Well-established methodologies were used to generate signals from each source. The study focused on ADRs related to three high-profile serious adverse reactions. A reference standard of over 600 established and plausible ADRs was created and used to evaluate the proposed approach against a comparator. RESULTS: The combined signaling system achieved a statistically significant large improvement over AERS (baseline) in the precision of top ranked signals. The average improvement ranged from 31% to almost threefold for different evaluation categories. Using this system, we identified a new association between the agent, rasburicase, and the adverse event, acute pancreatitis, which was supported by clinical review. CONCLUSIONS: The results provide promising initial evidence that combining AERS with EHRs via the framework of replicated signaling can improve the accuracy of signal detection for certain operating scenarios. The use of additional EHR data is required to further evaluate the capacity and limits of this system and to extend the generalizability of these results.

Facilitating adverse drug event detection in pharmacovigilance databases using molecular structure similarity: application to rhabdomyolysis.

BACKGROUND: Adverse drug events (ADE) cause considerable harm to patients, and consequently their detection is critical for patient safety. The US Food and Drug Administration maintains an adverse event reporting system (AERS) to facilitate the detection of ADE in drugs. Various data mining approaches have been developed that use AERS to detect signals identifying associations between drugs and ADE. The signals must then be monitored further by domain experts, which is a time-consuming task. OBJECTIVE: To develop a new methodology that combines existing data mining algorithms with chemical information by analysis of molecular fingerprints to enhance initial ADE signals generated from AERS, and to provide a decision support mechanism to facilitate the identification of novel adverse events. RESULTS: The method achieved a significant improvement in precision in identifying known ADE, and a more than twofold signal enhancement when applied to the ADE rhabdomyolysis. The simplicity of the method assists in highlighting the etiology of the ADE by identifying structurally similar drugs. A set of drugs with strong evidence from both AERS and molecular fingerprint-based modeling is constructed for further analysis. CONCLUSION: The results demonstrate that the proposed methodology could be used as a pharmacovigilance decision support tool to facilitate ADE detection.

Mining multi-item drug adverse effect associations in spontaneous reporting systems.

BACKGROUND: Multi-item adverse drug event (ADE) associations are associations relating multiple drugs to possibly multiple adverse events. The current standard in pharmacovigilance is bivariate association analysis, where each single drug-adverse effect combination is studied separately. The importance and difficulty in the detection of multi-item ADE associations was noted in several prominent pharmacovigilance studies. In this paper we examine the application of a well established data mining method known as association rule mining, which we tailored to the above problem, and demonstrate its value. The method was applied to the FDAs spontaneous adverse event reporting system (AERS) with minimal restrictions and expectations on its output, an experiment that has not been previously done on the scale and generality proposed in this work. RESULTS: Based on a set of 162,744 reports of suspected ADEs reported to AERS and published in the year 2008, our method identified 1167 multi-item ADE associations. A taxonomy that characterizes the associations was developed based on a representative sample. A significant number (67% of the total) of potential multi-item ADE associations identified were characterized and clinically validated by a domain expert as previously recognized ADE associations. Several potentially novel ADEs were also identified. A smaller proportion (4%) of associations were characterized and validated as known drug-drug interactions. CONCLUSIONS: Our findings demonstrate that multi-item ADEs are present and can be extracted from the FDA's adverse effect reporting system using our methodology, suggesting that our method is a valid approach for the initial identification of multi-item ADEs. The study also revealed several limitations and challenges that can be attributed to both the method and quality of data.

Under-documentation of chronic kidney disease in the electronic health record in outpatients.

OBJECTIVE: To ascertain if outpatients with moderate chronic kidney disease (CKD) had their condition documented in their notes in the electronic health record (EHR). DESIGN: Outpatients with CKD were selected based on a reduced estimated glomerular filtration rate and their notes extracted from the Columbia University data warehouse. Two lexical-based classification tools (classifier and word-counter) were developed to identify documentation of CKD in electronic notes. MEASUREMENTS: The tools categorized patients' individual notes on the basis of the presence of CKD-related terms. Patients were categorized as appropriately documented if their notes contained reference to CKD when CKD was present. RESULTS: The sensitivities of the classifier and word-count methods were 95.4% and 99.8%, respectively. The specificity of both was 99.8%. Categorization of individual patients as appropriately documented was 96.9% accurate. Of 107 patients with manually verified moderate CKD, 32 (22%) lacked appropriate documentation. Patients whose CKD had not been appropriately documented were significantly less likely to be on renin-angiotensin system inhibitors or have urine protein quantified, and had the illness for half as long (15.1 vs 30.7 months; p<0.01) compared to patients with documentation. CONCLUSION: Our studies show that lexical-based classification tools can accurately ascertain if appropriate documentation of CKD is present in a EHR. Using this method, we demonstrated under-documentation of patients with moderate CKD. Under-documented patients were less likely to receive CKD guideline recommended care. A tool that prompts providers to document CKD might shorten the time to implementing guideline-based recommendations.

Integrating heterogeneous knowledge sources to acquire executable drug-related knowledge.

Knowledge of medical entities, such as drug-related information is critical for many automated biomedical applications, such as decision support and pharmacovigilance. In this work, heterogeneous information sources were integrated automatically to obtain drug-related knowledge. We focus on one type of knowledge, drug-treats-condition, in the study and propose a framework for integrating disparate knowledge sources. Evaluation based on a random sample of drug-condition pairs indicated an overall coverage of 96%, recall of 98% and a precision of 87%. In conclusion, the preliminary study demonstrated that the knowledge generated from this study was comparable to the manually curated gold standard and that this method of automatically integrating knowledge sources is effective. The automated method should also be applicable to integrate other clinical knowledge, such as drug-related knowledge with omics information.

Aligning Structured and Unstructured Medical Problems Using UMLS.

This paper reports a pilot study to align medical problems in structured and unstructured EHR data using UMLS. A total of 120 medical problems in discharge summaries were extracted using NLP software (MedLEE) and aligned with 87 ICD-9 diagnoses for 19 non-overlapping hospital visits of 4 patients. The alignment accuracy was evaluated by a medical doctor. The average overlap of medical problems between the two data sources obtained by our automatic alignment method was 23.8%, which was about half of the manual review result, 43.56%. We discuss the implications for related research in integrating structured and unstructured EHR data.

Statistical Mining of Potential Drug Interaction Adverse Effects in FDA's Spontaneous Reporting System.

Many adverse drug effects (ADEs) can be attributed to drug interactions. Spontaneous reporting systems (SRS) provide a rich opportunity to detect novel post-marketed drug interaction adverse effects (DIAEs), as they include populations not well represented in clinical trials. However, their identification in SRS is nontrivial. Most existing research have addressed the statistical issues used to test or verify DIAEs, but not their identification as part of a systematic large scale database-wide mining process as discussed in this work. This paper examines the application of a highly optimized and tailored implementation of the Apriori algorithm, as well as methods addressing data quality issues, to the identification of DIAEs in FDAs SRS.

Voice capture of medical residents' clinical information needs during an inpatient rotation.

OBJECTIVE: To identify some of the challenges that medical residents face in addressing their information needs in an inpatient setting, by examining how voice capture in natural language of clinical questions fits into workflow, and by characterizing the focus, format, and semantic content and complexity of their questions. DESIGN: Internal medicine residents captured information needs on a digital recorder while on a hospital inpatient service and then participated in semi-structured interviews. MEASUREMENTS: Interviews were analyzed to identify emergent themes. Recorded questions were analyzed for focus (diagnosis, treatment, or epidemiology) and format, either foreground (specific knowledge relating to an individual patient) or background (general knowledge about a condition). Semantic concepts and types were identified using MetaMap (UMLS - Unified Medical Language System) and manually. RESULTS: Voice recording of questions appeared to unmask residents' latent information needs. Although residents were able to record questions during workflow, there was a delay from the time questions materialized to when they were recorded. Question focus was distributed among diagnosis (32%), treatment (40%), and epidemiology (28%), and the majority of questions were background (69%). Questions were semantically complex; foreground and background questions averaged 12.6 (SD 6.0) and 9.1 (SD 6.0) UMLS concepts, respectively. MetaMap failed to recognize concepts when residents used acronyms or abbreviations or omitted key terms. CONCLUSIONS: We found that it is feasible for residents to capture their clinical questions in natural language during workflow and that recording questions may prompt awareness of previously unrecognized information needs. However, the semantic complexity of typical questions and mapping failures due to residents' use of acronyms and abbreviations present challenges to machine-based extraction of semantic content.

Comparing ICD9-encoded diagnoses and NLP-processed discharge summaries for clinical trials pre-screening: a case study.

The prevalence of electronic medical record (EMR) systems has made mass-screening for clinical trials viable through secondary uses of clinical data, which often exist in both structured and free text formats. The tradeoffs of using information in either data format for clinical trials screening are understudied. This paper compares the results of clinical trial eligibility queries over ICD9-encoded diagnoses and NLP-processed textual discharge summaries. The strengths and weaknesses of both data sources are summarized along the following dimensions: information completeness, expressiveness, code granularity, and accuracy of temporal information. We conclude that NLP-processed patient reports supplement important information for eligibility screening and should be used in combination with structured data.