Extending the boundaries of cancer therapeutic complexity with literature text mining.

Drug combination therapy is a main pillar of cancer therapy. As the number of possible drug candidates for combinations grows, the development of optimal high complexity combination therapies (involving 4 or more drugs per treatment) such as RCHOP-I and FOLFIRINOX becomes increasingly challenging due to combinatorial explosion. In this paper, we propose a text mining (TM) based tool and workflow for rapid generation of high complexity combination treatments (HCCT) in order to extend the boundaries of complexity in cancer treatments. Our primary objectives were: (1) Characterize the existing limitations in combination therapy; (2) Develop and introduce the Plan Builder (PB) to utilize existing literature for drug combination effectively; (3) Evaluate PB's potential in accelerating the development of HCCT plans. Our results demonstrate that researchers and experts using PB are able to create HCCT plans at much greater speed and quality compared to conventional methods. By releasing PB, we hope to enable more researchers to engage with HCCT planning and demonstrate its clinical efficacy.

From centralized to ad-hoc knowledge base construction for hypotheses generation.

OBJECTIVE: To demonstrate and develop an approach enabling individual researchers or small teams to create their own ad-hoc, lightweight knowledge bases tailored for specialized scientific interests, using text-mining over scientific literature, and demonstrate the effectiveness of these knowledge bases in hypothesis generation and literature-based discovery (LBD). METHODS: We propose a lightweight process using an extractive search framework to create ad-hoc knowledge bases, which require minimal training and no background in bio-curation or computer science. These knowledge bases are particularly effective for LBD and hypothesis generation using Swanson's ABC method. The personalized nature of the knowledge bases allows for a somewhat higher level of noise than "public facing" ones, as researchers are expected to have prior domain experience to separate signal from noise. Fact verification is shifted from exhaustive verification of the knowledge base to post-hoc verification of specific entries of interest, allowing researchers to assess the correctness of relevant knowledge base entries by considering the paragraphs in which the facts were introduced. RESULTS: We demonstrate the methodology by constructing several knowledge bases of different kinds: three knowledge bases that support lab-internal hypothesis generation: Drug Delivery to Ovarian Tumors (DDOT); Tissue Engineering and Regeneration; Challenges in Cancer Research; and an additional comprehensive, accurate knowledge base designated as a public resource for the wider community on the topic of Cell Specific Drug Delivery (CSDD). In each case, we show the design and construction process, along with relevant visualizations for data exploration, and hypothesis generation. For CSDD and DDOT we also show meta-analysis, human evaluation, and in vitro experimental evaluation. CONCLUSION: Our approach enables researchers to create personalized, lightweight knowledge bases for specialized scientific interests, effectively facilitating hypothesis generation and literature-based discovery (LBD). By shifting fact verification efforts to post-hoc verification of specific entries, researchers can focus on exploring and generating hypotheses based on their expertise. The constructed knowledge bases demonstrate the versatility and adaptability of our approach to versatile research interests. The web-based platform, available at https://spike-kbc.apps.allenai.org, provides researchers with a valuable tool for rapid construction of knowledge bases tailored to their needs.

Using topic models to identify clients' functioning levels and alliance ruptures in psychotherapy.

Computerized natural language processing techniques can analyze psychotherapy sessions as texts, thus generating information about the therapy process and outcome and supporting the scaling-up of psychotherapy research. We used topic modeling to identify topics discussed in psychotherapy sessions and explored (a) which topics best identified clients' functioning and alliance ruptures and (b) whether changes in these topics were associated with changes in outcome. Transcripts of 873 sessions from 58 clients treated by 52 therapists were analyzed. Before each session, clients self-reported functioning and symptom level. After each session, therapists reported the extent of alliance rupture. Latent Dirichlet allocation was used to extract latent topics from psychotherapy textual data. Then a sparse multinomial logistic regression model was used to predict which topics best identified clients' functioning levels and the occurrence of alliance ruptures in psychotherapy sessions. Finally, we used multilevel growth models to explore the associations between changes in topics and changes in outcome. Session-based processing yielded a list of semantic topics. The model identified the labels above chance (65% to 75% accuracy). Change trajectories in topics were associated with change trajectories in outcome. The results suggest that topic models can exploit rich linguistic data within sessions to identify psychotherapy process and outcomes. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Using computerized text analysis to examine associations between linguistic features and clients' distress during psychotherapy.

Raw linguistic data within psychotherapy sessions may provide important information about clients' progress and well-being. In the current study, computerized text analytic techniques were applied to examine whether linguistic features were associated with clients' experiences of distress within and between clients and whether changes in linguistic features were associated with changes in treatment outcome. Transcripts of 729 psychotherapy sessions from 58 clients treated by 52 therapists were analyzed. Prior to each session, clients reported their distress level. Linguistic features were extracted automatically by using natural language parser for first-person singular identification and using positive and negative emotion words lexicon. The association between linguistic features and levels of distress was examined using multilevel models. At the within-client level, fewer first-person singular words, fewer negative emotional words and more positive emotional words were associated with lower distress in the same session; and fewer negative emotion words were associated with lower next session distress (rather small f2 effect sizes = 0.011 < f2 < 0.022). At the between-client level, only first session use of positive emotion words was associated with first session distress (ηp2 effect size = 0.08). A drop in the use of first-person singular words was associated with improved outcome from pre- to posttreatment (small ηp2 effect size = 0.05). The findings provide preliminary support for the association between clients' linguistic features and their fluctuating experience of distress. They point to the potential value of computerized linguistic measures to track therapeutic outcomes. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Mining fall-related information in clinical notes: Comparison of rule-based and novel word embedding-based machine learning approaches.

BACKGROUND: Natural language processing (NLP) of health-related data is still an expertise demanding, and resource expensive process. We created a novel, open source rapid clinical text mining system called NimbleMiner. NimbleMiner combines several machine learning techniques (word embedding models and positive only labels learning) to facilitate the process in which a human rapidly performs text mining of clinical narratives, while being aided by the machine learning components. OBJECTIVE: This manuscript describes the general system architecture and user Interface and presents results of a case study aimed at classifying fall-related information (including fall history, fall prevention interventions, and fall risk) in homecare visit notes. METHODS: We extracted a corpus of homecare visit notes (n = 1,149,586) for 89,459 patients from a large US-based homecare agency. We used a gold standard testing dataset of 750 notes annotated by two human reviewers to compare the NimbleMiner's ability to classify documents regarding whether they contain fall-related information with a previously developed rule-based NLP system. RESULTS: NimbleMiner outperformed the rule-based system in almost all domains. The overall F- score was 85.8% compared to 81% by the rule based-system with the best performance for identifying general fall history (F = 89% vs. F = 85.1% rule-based), followed by fall risk (F = 87% vs. F = 78.7% rule-based), fall prevention interventions (F = 88.1% vs. F = 78.2% rule-based) and fall within 2 days of the note date (F = 83.1% vs. F = 80.6% rule-based). The rule-based system achieved slightly better performance for fall within 2 weeks of the note date (F = 81.9% vs. F = 84% rule-based). DISCUSSION & CONCLUSIONS: NimbleMiner outperformed other systems aimed at fall information classification, including our previously developed rule-based approach. These promising results indicate that clinical text mining can be implemented without the need for large labeled datasets necessary for other types of machine learning. This is critical for domains with little NLP developments, like nursing or allied health professions.

Rich parameterization improves RNA structure prediction.

Current approaches to RNA structure prediction range from physics-based methods, which rely on thousands of experimentally measured thermodynamic parameters, to machine-learning (ML) techniques. While the methods for parameter estimation are successfully shifting toward ML-based approaches, the model parameterizations so far remained fairly constant. We study the potential contribution of increasing the amount of information utilized by RNA folding prediction models to the improvement of their prediction quality. This is achieved by proposing novel models, which refine previous ones by examining more types of structural elements, and larger sequential contexts for these elements. Our proposed fine-grained models are made practical thanks to the availability of large training sets, advances in machine-learning, and recent accelerations to RNA folding algorithms. We show that the application of more detailed models indeed improves prediction quality, while the corresponding running time of the folding algorithm remains fast. An additional important outcome of this experiment is a new RNA folding prediction model (coupled with a freely available implementation), which results in a significantly higher prediction quality than that of previous models. This final model has about 70,000 free parameters, several orders of magnitude more than previous models. Being trained and tested over the same comprehensive data sets, our model achieves a score of 84% according to the F1-measure over correctly-predicted base-pairs (i.e., 16% error rate), compared to the previously best reported score of 70% (i.e., 30% error rate). That is, the new model yields an error reduction of about 50%. Trained models and source code are available at www.cs.bgu.ac.il/?negevcb/contextfold.