Marco Dominietto
marcodominietto.bsky.social
Marco Dominietto
@marcodominietto.bsky.social
Recent surveys suggest that around 60% of US adults would be uncomfortable with their physician relying on #AI. Another 63% would want to be informed if it were used in their care, making it clear that ethical guidelines are needed to ensure patients are kept in the loop.

#InformedConsent #Ethics
Ethical Obligations to Inform Patients About Use of AI Tools
This Perspective discusses the use of artificial intelligence (AI) tools in health care and offers a framework for deciding what patients should be told about AI tools.
jamanetwork.com
August 11, 2025 at 8:16 AM
For investors, #AI is no longer a differentiator.

AI-related investments now account for 71% of all VC funding in the US, meaning only those whose use of AI makes a meaningful difference will stand out. Now is the time to think critically about AI-related ideas.

#VentureCapital #Startups
August 8, 2025 at 8:18 AM
There will be times when someone asks a #question we can't answer. Sometimes there will be no answer.

That doesn't mean we can't still provide #clarity and certainty to our teams. By acknowledging knowledge gaps and prioritising 'good enough' answers, we can give the support they need.

#Leadership
August 7, 2025 at 8:15 AM
Just 10% of #PancreaticCancer cases can be explained by genetic or familial factors, making prediction tricky.

However, a new study suggests that using #AI to analyse EHR could make it easier to pick up on patients without traditional risk factors or genetic variants.

#Oncology
Enhancing EHR-based pancreatic cancer prediction with LLM-derived embeddings - npj Digital Medicine
Pancreatic cancer (PC) is often diagnosed late, as early symptoms and effective screening tools are lacking, and genetic or familial factors explain only ~10% of cases. Leveraging longitudinal electronic health record (EHR) data may offer a promising avenue for early detection. We developed a predictive model using large language model (LLM)-derived embeddings of medical condition names to enhance
www.nature.com
August 7, 2025 at 7:56 AM
Research and business are both uncertain. That doesn't mean you can't keep moving forward.

Making decisions in an uncertain environment starts with focusing on what you are certain about.

#Uncertainty #DecisionMaking #Leadership
August 6, 2025 at 8:52 AM
To better benchmark #AI in #healthcare, researchers have proposed using a FeTS Challenge that brings together data from multiple sites.

It's an interesting approach, but more importantly, it reflects the growing importance of aligning digital tools with patients' realities.

#FederatedLearning
Towards fair decentralized benchmarking of healthcare AI algorithms with the Federated Tumor Segmentation (FeTS) challenge - PubMed
Computational competitions are the standard for benchmarking medical image analysis algorithms, but they typically use small curated test datasets acquired at a few centers, leaving a gap to the reality of diverse multicentric patient data. To this end, the Federated Tumor Segmentation (FeTS) Challe …
pubmed.ncbi.nlm.nih.gov
August 6, 2025 at 8:17 AM
The first step is always the hardest.

To motivate employees to follow through on their ideas, stop focusing on the finished product. Instead, focus on encouraging them to make a start.

It doesn't have to be perfect; it just has to be enough to build off of.
July 27, 2025 at 8:18 AM
Knowing when to kill a doomed project starts with building a culture of #respect. By respecting others and the resources given to us, we can let go of failing projects more easily.

#ProjectManagement #Mindset
July 26, 2025 at 8:17 AM
Catching #FattyLiverDisease before it causes #LiverCancer can be expensive.

However, a new AI tool opens up the possibility of using X-rays to enable cheaper and faster detection.

Source: https://pubs.rsna.org/doi/10.1148/ryct.240402

#AI
July 25, 2025 at 8:17 AM
#AIScreening for breast cancer has matched the accuracy of radiologists and reduced workload by 44%. A Swedish study analysing over 80,000 mammograms showed AI and human readers detected #BreastCancer at similar rates, but AI drastically cut reading time.

https://tinyurl.com/yc2txfk7
July 25, 2025 at 8:15 AM
#PersonalisedCare is no longer an “emerging idea”; it’s the standard in many medical fields.

#TailoredTreatment plans reduce risk, improve outcomes & help streamline healthcare costs. A one-size-fits-all approach is not inclusive in a medical setting. We need more targeted, personalised care.
July 24, 2025 at 8:15 AM
Developing a genius solution is only half the battle.

For #startups to succeed, they need to use their idea effectively. With this in mind, investors should pay close attention to a startup's #leadership and whether they have the skills to unlock a solution's full potential.

#VentureCapital
July 24, 2025 at 8:15 AM
A recent assessment of an #MMAI algorithm designed for use in #ProstateCancer prognostics found no evidence of algorithmic #bias.

Whilst this only focused on one tool, it's a valuable reminder that we need to regularly test AI tools for bias.

#AI
Assessing Algorithmic Fairness With a Multimodal Artificial Intelligence Model in Men of African and Non-African Origin on NRG Oncology Prostate Cancer Phase III Trials - PubMed
Using cooperative group data sets with a racially diverse population, the MMAI algorithm performed well across racial subgroups without evidence of algorithmic bias.
pubmed.ncbi.nlm.nih.gov
July 23, 2025 at 8:18 AM
#EthicalLeadership is easy when people agree with you; it's a little more difficult when they don't.

However, disagreement is essential. Without it, how can we consider new perspectives? How can we be sure we're doing the right thing?

We need to embrace #disagreement, not fear it.

#Leadership
July 11, 2025 at 8:17 AM
Explainable #MachineLearning tools in #oncology aren't just important because they improve confidence in decision-making. They're also important for helping clinicians learn by giving them insight into why a tool generates a particular output.

#Learning
July 11, 2025 at 8:16 AM
A new study, which found that an AI-based prognostic test can effectively predict clinical risk in post-prostatectomy settings, makes me wonder how #AI will change #prognostics.

What changes will we need to make to ensure it's effectively embedded into clinical practices?

#ProstateCancer
Artificial Intelligence–Based Digital Histologic Classifier for Prostate Cancer Risk Stratification: Independent Blinded Validation in Patients Treated With Radical Prostatectomy | JCO Clinical Cancer Informatics
PURPOSEArtificial intelligence (AI) tools that identify pathologic features from digitized whole-slide images (WSIs) of prostate cancer (CaP) generate data to predict outcomes. The objective of this study was to evaluate the clinical validity of an AI-...
ascopubs.org
July 10, 2025 at 8:18 AM
Looking forward to hosting the Health and Science Tech Community group meeting later today!

It's always great to come together with #HealthTech founders to discuss the industry and the challenges their #startups are facing.

Hope to see you there!

#Networking
Event: Health and Science Tech Community with Marco Dominietto on Jul 9, 2025
The HealthTech Community brings together founders in healthcare tech, biotech, life sciences, digital health, and more to tackle the challenges that healthcare startups face. This is part of a cohort breakout.
www.startups.com
July 9, 2025 at 10:02 AM
Whether we're trying to scale a business or simply improve our #TimeManagement, we should focus on optimising rather than maximising.

When we maximise, we leave no room for the unexpected. As a result, we may miss out on new opportunities for #growth and transformation.

#Scaling
July 2, 2025 at 8:15 AM
Our first #strategy isn't going to work forever.

These three questions can help us understand whether a #change is needed:

1. Is our current strategy still getting good results?
2. Could these results be better?
3. Could we do things differently? If so, how?

#BusinessManagement #Growth
July 2, 2025 at 8:09 AM
Scientifically validated #AI tools are heralded as the beginning of a new era of #oncology, but how do they stack up against traditional diagnostic approaches for #ProstateCancer?

A recent study suggests that AI-assisted systems could prove to be superior.

#CancerDiagnostics
AI-Assisted vs Unassisted Identification of Prostate Cancer in Magnetic Resonance Images
Is the use of a scientifically validated artificial intelligence (AI) system associated with improved diagnostic accuracy for detecting clinically significant prostate cancer (csPCa) on magnetic resonance imaging compared with unassisted readings? ...
pmc.ncbi.nlm.nih.gov
June 30, 2025 at 9:10 AM
#Urgency vs perfection: which matters more?

Slow decision making leads to slower #growth, making it crucial for leaders to overcome #LeadershipInertia. It's time to start asking: Are we waiting when we should be moving?

#Leadership
June 27, 2025 at 8:36 AM
#AI is only as good as the #data it's trained on, which means it's time to prioritise #DataManagement.

Harmonising our data will help to reduce noise in AI models and enable more meaningful outputs for individual patients, paving the way for better care and outcomes.

#Oncology #FutureOfHealthcare
June 26, 2025 at 8:31 AM
#AI hype is at an all-time high. For investors, that means it's time to prioritise #quality over quantity by questioning whether a startup's use of AI really creates a competitive advantage.

#HealthTech #VentureCapital
June 25, 2025 at 8:16 AM
#AI was recently used in a closed-loop system to help scientists identify potential drug combos for use in #BreastCancer treatment, and it got me thinking: as AI changes the way we work, how does our relationship with it need to change?

#AICollaboration
June 24, 2025 at 8:45 AM
#ProstateCancer is personal; treatment needs to be, too.

A new AI-powered test has shown potential in identifying which high-risk patients are more likely to respond well to abiraterone, providing us with a better understanding of how AI can be used to personalise treatment plans.

#AI #Oncology
External validation of a digital pathology-based multimodal artificial intelligence-derived prognostic model in patients with advanced prostate cancer starting long-term androgen deprivation therapy: a post-hoc ancillary biomarker study of four phase 3 randomised controlled trials of the STAMPEDE platform protocol
Effective prognostication improves selection of patients with prostate cancer for treatment combinations. We aimed to evaluate whether a previously de…
www.sciencedirect.com
June 24, 2025 at 8:16 AM