Revolution in the pharmaceutical industry – NetraMark beats ChatGPT by a mile

Precision instead of a black box

NetraMark develops AI systems specifically designed for the pharmaceutical industry. At its core is the NetraAI platform, whose algorithmic basis differs from classic machine learning approaches and generic language models. Instead of relying on large, unstructured data sets, NetraAI also processes smaller, highly complex clinical data sets with maximum explainability.

The secret lies in a meta-evolutionary framework that combines principles of dynamic systems theory, information geometry, and evolutionary computation. Using only 2 to 4 explainable variables, NetraAI discovers so-called "personas," - clinically relevant patient subgroups. These groups are statistically significant and directly applicable for clinical study design and regulatory drug approval processes. The AI agent "Dr. Netra" contributes additional expertise from medical literature without compromising interpretability.

This technology enables NetraMark to gain valid insights even with small data sets and classify patient subtypes that respond particularly well or poorly to certain drugs. This is an enormous advantage for study planning, precision medicine, and regulatory approvals.

NetraAI clearly outperforms ChatGPT and DeepSeek

In a recent comparative study, NetraMark tested its platform against ChatGPT, DeepSeek, and traditional machine learning methods. The study was based on three renowned and highly complex clinical datasets: CATIE (schizophrenia), CAN-BIND (depression), and COMPASS (pancreatic cancer). The results were groundbreaking, as only NetraAI was able to generate clinically meaningful and statistically valid insights.

In the CATIE dataset, NetraAI accurately identified patient groups with high response rates to medication. Model accuracy increased from 55–66% (ML standard) to over 85%. ChatGPT and DeepSeek, on the other hand, only provided generic statements with no clinical value. Neither a valid cluster nor practical interpretations could be found.

With CAN-BIND, NetraAI achieved an almost perfect result with AUC values of 0.99. The subgroups identified showed significant differences in response to treatment for depression. ChatGPT and DeepSeek were again unable to recognize meaningful patterns despite the reduced complexity of the data. Traditional ML remained stuck at 66% accuracy.

In the COMPASS dataset for pancreatic cancer, NetraAI achieved over 90% model accuracy even with highly heterogeneous cancer data, identified survival-relevant patient groups, and offered a new approach to treatment decisions. DeepSeek and ChatGPT delivered results at random levels that are unsuitable for oncology and precision medicine. Conventional algorithms failed in terms of consistency and interpretability.

In a class of its own

What makes NetraAI so special is its unique combination of explainability, robustness, and precision. While other models always predict something, NetraAI clearly distinguishes between highly predictable patient subgroups and those with higher uncertainty. This capability helps reduce misdiagnoses, placebo effects, and clinical trial failures.

Particularly noteworthy is its ability to work with a small number of highly relevant variables. This enables precise study submission criteria, faster recruitment of suitable test subjects, and strategically optimized study designs. In a market where up to 90% of clinical trials fail, NetraAI could make a decisive difference.

Chart before outbreak

With the pre-market announcement of a major milestone, the NetraMark share price was able to break out of the downward trend established since mid-April. Both the MACD trend-following indicator and the RSI issued buy signals. The next target is the horizontal resistance level at CAD 1.66.

The publication of the comparative study on arXiv marks a turning point in the history of NetraMark. The trend is moving away from generic, "all-powerful" AI systems toward highly specialized, medically validated multi-agent AI platforms. NetraMark could position itself as a first mover with a deep technological advantage in a market where reliability outweighs scale. Pharmaceutical companies stand to benefit in multiple ways, which could make NetraMark a hot takeover candidate in the future. In addition to cost reductions through fewer failures, big pharma also benefits from time savings and regulatory advantages thanks to robust, explainable models with transparent study designs. The capital market is likely to reward this sooner or later, especially since the findings are both regulatory-relevant and economically scalable. With a market capitalization of CAD 114.19 million, NetraMark still holds enormous upside potential.