Text Mining & NLP
To improve NLP analysis, Keras360 has formulated a lexicon library specifically designed for clinical research. The lexicons are systematically organized by pertinent topics. Please refer to the secondary lexicons, also referred to as the topics of clinical research, that your team intends to employ for Delphi-NLP sentiment analysis.
About the search
Since R1 in year 2024, Keras360 has conducted a thorough examination of scientific literature accrued over a minimum of five years up to the extraction date. In addition to presenting the total number of unique lexicons extracted, the last column of each table also contains lexicons classified according to non-neutrality weights for NLP. Our searches are predicated on our secondary lexicons and their corresponding customized search terms. For further information regarding the clinical lexicon library, please contact us via email at: contact.general@keras360.io
Main Library, Family Lexicon
Clinical, Economics, Computer Science, Biomedical Engineering
Latest extraction
September 1, 2025 (Round, R3)
Round 3 (R3)
In R3, 12 new secondary lexicons have been incorporated into the search, as indicated by the terms emphasized in bold; specifically, the rows spanning from CAR-T to cardiovascular. Results were based on Randomized Controlled Trials (RCT) and Clinical Trials with the exception for “Disease Diagnosis” search criteria.
Keras360 Clinical Lexicon Library currently holds 41 secondary lexicons aka. themes. Table below is also available in pdf.
| Secondary Lexicons | Unique Lexicons | Non-neutral Lexicons |
| CAR-T | 18,457 | 6,856 |
| insurance | claims | 147,880 | 54,578 |
| dental | oral health | 943,233 | 318,882 |
| osseointegration | 150,319 | 49,669 |
| radiotherapy | 126,085 | 43,919 |
| rare diseases | 32,564 | 11,281 |
| lymphedema | 18,549 | 6,716 |
| imaging scans | 988,788 | 326,210 |
| palliative care | 90,203 | 35,842 |
| aging | senescence | 214,333 | 77,674 |
| bridging | consolidative therapy | 8,235 | 2,982 |
| cardiovascular | 411,111 | 143,194 |
| transplant | 424,914 | 148,151 |
| disease.severity | 535,926 | 193,310 |
| pain | 1,053,362 | 374,892 |
| adverse.event | 865,867 | 298,560 |
| dose.response | 594,348 | 200,959 |
| waiting.list | 51,183 | 20,463 |
| switch.regimen | 24,887 | 8,634 |
| vaccination | 268,748 | 90,829 |
| AI | 117,626 | 42,638 |
| neurodegenerative | 157,109 | 54,916 |
| Gastrointestinal (GI) | 370,333 | 127,001 |
| chemotherapy | immunotherapy | 530,377 | 187,558 |
| cost benefit | 232,289 | 85,360 |
| cost utility | 33,207 | 12,226 |
| ocular | 41,574 | 14,210 |
| disease.diagnosis | 1,868,281 | 647,889 |
| STD | 304,013 | 105,211 |
| reproductive | 192,250 | 64,336 |
| family.planning | 46,636 | 15,708 |
| renal.failure | 707,734 | 245,815 |
| nutrition | 771,118 | 270,493 |
| obesity | 423,083 | 150,169 |
| personality.disorder | 33,245 | 13,662 |
| precision.medicine | 185,131 | 71,840 |
| psychology | 68,792 | 28,281 |
| reconstructive | 112,715 | 39,808 |
| QOL | 1,150,468 | 432,305 |
| statistics | 580,666 | 206,474 |
| wounds | 409,917 | 143,327 |
| Total | 15,305,556 | 5,372,828 |
N.B., In this tabulation, “introns” lexicons and duplicates that emerged during the extraction have been removed – one of the rationales for the notable variations in unique lexicons from R2 to R3.
Video NLP
a) Xenotransplantation presents a crucial life-saving opportunity. While some perspectives lean towards agreement, they often remain neutral in their stance. For instance, certain viewpoints such as opinion 2 shown in the video below suggests that sentiment scores may be misleading without taking into account the weights associated with clinical lexicon; thus, the inclusion of a Medical Lexicon is essential.
b) Semaglutide: Drug123 (pseudo-name/anonymised) is indicated for the management of weight and type-2 diabetic treatment. This medication has been demonstrated to yield effective outcomes in weight reduction and glycemic control. Nevertheless, it is important to note that it carries a rare risk of sudden vision loss associated with Non-Arteritic Anterior Ischemic Optic Neuropathy (NAION).
Videos made by: Keras360, The Waterhouse SG
Music by: Alexandre Desplat
Previous extraction
Please select each tab to gain further insight into the fundamental descriptive statistics of the extracted lexicons.
Round 1 (R1) : May 4, 2024
| Secondary Lexicons | Unique Lexicons | Non-neutral Lexicons |
Transplant Disease.severity Pain Adverse.event Dose.response Waiting.list Switch.regimen Vaccination AI Neurodegen GI Chemoimmuno Cost.benefit Cost.utility Ocular | 384,581 532,105 942,474 839,166 570,375 47,582 22,423 241,777 88,963 152,327 328,008 507,577 214,169 29,823 18,746 | 133,305 191,953 332,805 289,334 192,569 19,048 7,788 81,002 31,264 53,269 111,326 178,471 78,564 10,958 6,355 |
| Total | 4,920,096 | 1,718,011 |
Round 2 (R2) : January 5, 2025
| Secondary Lexicon | Unique | Non-Neutral |
|---|---|---|
| Round 1 + 2 | ||
| Transplant | Transplantation | 430,880 | 149,574 |
| Disease Severity | 538,681 | 194,410 |
| Pain | 1,091,408 | 385,810 |
| Adverse Event | 873,541 | 301,242 |
| Dose Response | 596,931 | 201,331 |
| Waiting List | 51,645 | 20,725 |
| Switch Regimen | 22,553 | 7,837 |
| Vaccination * (a) | 303,710 | 102,038 |
| AI * (a) | 133,910 | 47,383 |
| Neurodegenerative | 154,980 | 54,159 |
| Gastrointestinal | 380,397 | 129,614 |
| Chemotherapy | 584,032 | 205,813 |
| Cost Benefit | 237,418 | 87,217 |
| Cost Utility | 33,311 | 12,282 |
| Ocular | 35,361 | 11,928 |
| Round 2 | ||
| Renal Disease (a) | 672,112 | 232,691 |
| Nutrition | 728,251 | 254,012 |
| Obesity | Overweight | Binge | 400,990 | 141,851 |
| Reproductive* | 173,387 | 57,379 |
| STDs* | 287,635 | 99,324 |
| Family Planning | Contraceptives | 45,134 | 15,187 |
| Wounds | 387,061 | 134,465 |
| Precision Medicine | 164,889 | 63,449 |
| Statistics | Clinical Study Design | 553,781 | 195,988 |
| QoL | 1,069,522 | 400,279 |
| Reconstructive* | 105,142 | 36,867 |
| Personality Disorder* | 30,558 | 12,607 |
| Psychological Trauma | 64,571 | 26,331 |
| Disease Diagnosis | 1,831,181 | 629,965 |
| Total (R1+ R2 group) | 5,468,758 | 1,911,363 |
| Total (R2 group) | 6,514,214 | 2,300,395 |
| Total (all groups) | 11,982,972 | 4,211,758 |
* contains exclusive strings for search terms
(a) The extraction contained at least one corrupted file; this issue was resolved on January 6, 2025.
Results were based on Randomized Controlled Trials (RCT) and Clinical Trials with the exception for “Disease Diagnosis” search criteria.
Relevant topics:
