<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE article
  PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.0 20120330//EN" "http://jats.nlm.nih.gov/publishing/1.0/JATS-journalpublishing1.dtd">
<article article-type="letter" dtd-version="1.0" specific-use="sps-1.8" xml:lang="es" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">
	<front>
		<journal-meta>
			<journal-id journal-id-type="publisher-id">rac</journal-id>
			<journal-title-group>
				<journal-title>Revista argentina de cardiología</journal-title>
				<abbrev-journal-title abbrev-type="publisher">Rev Argent Cardiol</abbrev-journal-title>
			</journal-title-group>
			<issn pub-type="ppub">0034-7000</issn>
			<issn pub-type="epub">1850-3748</issn>
			<publisher>
				<publisher-name>Sociedad Argentina de Cardiología</publisher-name>
			</publisher>
		</journal-meta>
		<article-meta>
			<article-id pub-id-type="doi">10.7775/rac.es.v93.i2.20877</article-id>
			<article-id pub-id-type="publisher-id">00015</article-id>
			<article-categories>
				<subj-group subj-group-type="heading">
					<subject>CARTAS DE LECTORES</subject>
				</subj-group>
			</article-categories>
			<title-group>
				<article-title>“Doctor, ¿Me tengo que preocupar por esto?”</article-title>
				<trans-title-group xml:lang="en">
					<trans-title>“Doctor, Should I Be Worried About This?”</trans-title>
				</trans-title-group>
			</title-group>
			<contrib-group>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0001-5130-626X</contrib-id>
					<name>
						<surname>BATTIONI</surname>
						<given-names>LUCIANO</given-names>
					</name>
					<xref ref-type="aff" rid="aff1"><sup>1</sup></xref>
					<xref ref-type="fn" rid="fn1"><sup>MTSAC</sup></xref>
				</contrib>
				</contrib-group>
				<aff id="aff1">
					<label>1</label>
					<institution content-type="original"> Director del post grado de inteligencia artificial aplicado a ciencias de la salud. Universidad Nacional del Litoral. Santa Fe, Santa Fe</institution>
					<institution content-type="normalized">Universidad Nacional del Litoral</institution>
					<institution content-type="orgname">Universidad Nacional del Litoral</institution>
					<addr-line>
						<named-content content-type="city">Santa Fe</named-content>
					</addr-line>
					<country country="AR">Argentina</country>
					 <email>lucianobattioni@gmail.com</email>
				</aff>
			<author-notes>
				<corresp id="c1">
					<label> Dirección para correspondencia </label>: Luciano Battioni, Calle 15 N.° 669<italic>,</italic> Mercedes, Buenos Aires. E-mail: <email>lucianobattioni@gmail.com</email>
				</corresp>
				<fn fn-type="conflict" id="fn2">
					<label>Declaración de conflicto de intereses</label>
					<p> Los autores declaran que no tienen conflicto de intereses. (Ver formulario de conflicto de intereses en la web)</p>
				</fn>
			</author-notes>
			<!--<pub-date date-type="pub" publication-format="electronic">
				<day>16</day>
				<month>05</month>
				<year>2025</year>
			</pub-date>
			<pub-date date-type="collection" publication-format="electronic">
				<season>Mar-Apr</season>
				<year>2025</year>
			</pub-date>-->
			<pub-date pub-type="epub-ppub">
				<season>Mar-Apr</season>
				<year>2025</year>
			</pub-date>
			<volume>93</volume>
			<issue>2</issue>
			<fpage>164</fpage>
			<lpage>165</lpage>
			<permissions>
				<license license-type="open-access" xlink:href="https://creativecommons.org/licenses/by-nc/4.0/" xml:lang="es">
					<license-p>Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons</license-p>
				</license>
			</permissions>
			<counts>
				<fig-count count="0"/>
				<table-count count="0"/>
				<equation-count count="0"/>
				<ref-count count="5"/>
				<page-count count="2"/>
			</counts>
		</article-meta>
	</front>
	<body>
		<p>Si pudiéramos destilar los objetivos universales de las ciencias médicas, podríamos concluir en que son tres: diagnosticar, establecer un pronóstico y tratar.</p>
		<p>El peso relativo de cada una de ellos es distinto para el paciente y para el médico. Probablemente la pregunta con la que titulé esta carta sea la más importante para el paciente. Sin embargo, es la que menos desarrollo técnico científico ha tenido.</p>
		<p>En la práctica cotidiana utilizamos herramientas pronósticas de forma asidua y hasta dogmática; incluso, muchas veces, tratamos de utilizar <italic>scores</italic> generados para predecir un evento X en una población y extrapolarlos para un evento Y en otra. (<xref ref-type="bibr" rid="B1">1</xref>) La mayoría de estas herramientas tienen áreas bajo la curva ROC de entre 0,60 y 0,85. (<xref ref-type="bibr" rid="B2">2</xref>,<xref ref-type="bibr" rid="B3">3</xref>) Si ofreciéramos a alguien estas herramientas para detectar transacciones bancarias fraudulentas, rápidamente nos daría la mano en saludo y nos mostraría la salida. </p>
		<p>Este bajo rendimiento en la capacidad predictiva actual no solo se debe a múltiples limitantes y dificultades relacionadas con el manejo de datos en el campo de la salud, sino también a las herramientas hasta ahora utilizadas. En el trabajo titulado <italic>Capacidad predictiva de eventos en pacientes con hipertensión arterial mediante el análisis de redes neuronales artificiales del monitoreo ambulatorio de presión arterial en comparación con la estratificación de riesgo clínica,</italic> Di Gennaro y cols. desarrollaron un modelo de red neuronal sencillo que tiene la capacidad de predecir qué le pasará a nuestro paciente con mayor precisión. (<xref ref-type="bibr" rid="B4">4</xref>)</p>
		<p>Más allá de las limitaciones reconocidas por los autores, hay que destacar lo que este trabajo representa, la introducción de herramientas de inteligencia artificial (IA) al ejercicio clínico. La integración de la IA en medicina cambiará nuestra práctica en formas que no podemos vislumbrar todavía. Integrando múltiples variables, creando algunas que no conocíamos o relacionando hechos que se escapan al análisis humano, podremos ofrecer una medicina de precisión. (<xref ref-type="bibr" rid="B5">5</xref>)</p>
		<p>Pero no todo lo que brilla es oro. Por ejemplo, las redes neuronales tienden al <italic>overfitting</italic> (sobreajuste), es decir que tienen una gran validez interna, pero a la hora de ser validados en cohortes externas su capacidad puede caer significativamente.</p>
		<p>En conclusión, este trabajo representa una de las primeras instancias de uso de herramientas de IA en medicina a nivel nacional y, salvando sus limitaciones de diseño, nos da una pequeñísima muestra de lo que esta integración podría representar y nos insta a seguir investigando en este campo.</p>
	</body>
	<back>
		<ref-list>
			<title>BIBLIOGRAFÍA</title>
			<ref id="B1">
				<label>1</label>
				<mixed-citation>u JT, Wang SL, Chu YJ, Long DY, Dong JZ, Fan XW, et al. CHADS2 and CHA2DS2-VASc Scores Predict the Risk of Ischemic Stroke Outcome in Patients with Interatrial Block without Atrial Fibrillation. J Atheroscler Thromb 2017;24:176-84. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.5551/jat.34900">https://doi.org/10.5551/jat.34900</ext-link>
				</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Wu</surname>
							<given-names>JT</given-names>
						</name>
						<name>
							<surname>Wang</surname>
							<given-names>SL</given-names>
						</name>
						<name>
							<surname>Chu</surname>
							<given-names>YJ</given-names>
						</name>
						<name>
							<surname>Long</surname>
							<given-names>DY</given-names>
						</name>
						<name>
							<surname>Dong</surname>
							<given-names>JZ</given-names>
						</name>
						<name>
							<surname>Fan</surname>
							<given-names>XW</given-names>
						</name>
					</person-group>
					<article-title>CHADS2 and CHA2DS2-VASc Scores Predict the Risk of Ischemic Stroke Outcome in Patients with Interatrial Block without Atrial Fibrillation</article-title>
					<source>J Atheroscler Thromb</source>
					<year>2017</year>
					<volume>24</volume>
					<fpage>176</fpage>
					<lpage>184</lpage>
					<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.5551/jat.34900">https://doi.org/10.5551/jat.34900</ext-link>
				</element-citation>
			</ref>
			<ref id="B2">
				<label>2</label>
				<mixed-citation>&amp;apos;Agostino RB Sr, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008;117:743-53. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1161/CIRCULATIONAHA.107.699579">https://doi.org/10.1161/CIRCULATIONAHA.107.699579</ext-link>
				</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>D&amp;apos;Agostino</surname>
							<given-names>RB</given-names>
							<suffix>Sr</suffix>
						</name>
						<name>
							<surname>Vasan</surname>
							<given-names>RS</given-names>
						</name>
						<name>
							<surname>Pencina</surname>
							<given-names>MJ</given-names>
						</name>
						<name>
							<surname>Wolf</surname>
							<given-names>PA</given-names>
						</name>
						<name>
							<surname>Cobain</surname>
							<given-names>M</given-names>
						</name>
						<name>
							<surname>Massaro</surname>
							<given-names>JM</given-names>
						</name>
					</person-group>
					<article-title>General cardiovascular risk profile for use in primary care the Framingham Heart Study</article-title>
					<source>Circulation</source>
					<year>2008</year>
					<volume>117</volume>
					<fpage>743</fpage>
					<lpage>753</lpage>
					<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1161/CIRCULATIONAHA.107.699579">https://doi.org/10.1161/CIRCULATIONAHA.107.699579</ext-link>
				</element-citation>
			</ref>
			<ref id="B3">
				<label>3</label>
				<mixed-citation>Olesen JB, Lip GY, Hansen ML, Hansen PR, Tolstrup JS, Lindhardsen J, et al. Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study. BMJ 2011; 342:d124. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1136/bmj.d124">https://doi.org/10.1136/bmj.d124</ext-link>
				</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Olesen</surname>
							<given-names>JB</given-names>
						</name>
						<name>
							<surname>Lip</surname>
							<given-names>GY</given-names>
						</name>
						<name>
							<surname>Hansen</surname>
							<given-names>ML</given-names>
						</name>
						<name>
							<surname>Hansen</surname>
							<given-names>PR</given-names>
						</name>
						<name>
							<surname>Tolstrup</surname>
							<given-names>JS</given-names>
						</name>
						<name>
							<surname>Lindhardsen</surname>
							<given-names>J</given-names>
						</name>
						<etal/>
					</person-group>
					<article-title>Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study</article-title>
					<source>BMJ</source>
					<year>2011</year>
					<volume>342</volume>
					<fpage>d124</fpage>
					<lpage>d124</lpage>
					<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1136/bmj.d124">https://doi.org/10.1136/bmj.d124</ext-link>
				</element-citation>
			</ref>
			<ref id="B4">
				<label>4</label>
				<mixed-citation>Di Gennaro F P, Catalano MP, Aguirre AG, Fernández ML, Llanos R, Pérez Lloretet S, et al. Capacidad predictiva de eventos en pacientes con hipertensión arterial mediante el análisis con redes neuronales artificiales del monitoreo ambulatorio de presión arterial en comparación con la estratificación de riesgo clínica. Rev Argent Cardiol 2025;93:33-42. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7775/rac.es.v93.i1.20854">https://doi.org/10.7775/rac.es.v93.i1.20854</ext-link>
				</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Di Gennaro F</surname>
							<given-names>P</given-names>
						</name>
						<name>
							<surname>Catalano</surname>
							<given-names>MP</given-names>
						</name>
						<name>
							<surname>Aguirre</surname>
							<given-names>AG</given-names>
						</name>
						<name>
							<surname>Fernández</surname>
							<given-names>ML</given-names>
						</name>
						<name>
							<surname>Llanos</surname>
							<given-names>R</given-names>
						</name>
						<name>
							<surname>Pérez Lloretet</surname>
							<given-names>S</given-names>
						</name>
						<etal/>
					</person-group>
					<article-title>Capacidad predictiva de eventos en pacientes con hipertensión arterial mediante el análisis con redes neuronales artificiales del monitoreo ambulatorio de presión arterial en comparación con la estratificación de riesgo clínica</article-title>
					<source>Rev Argent Cardiol</source>
					<year>2025</year>
					<volume>93</volume>
					<fpage>33</fpage>
					<lpage>42</lpage>
					<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7775/rac.es.v93.i1.20854">https://doi.org/10.7775/rac.es.v93.i1.20854</ext-link>
				</element-citation>
			</ref>
			<ref id="B5">
				<label>5</label>
				<mixed-citation>Topol E. Deep medicine: How artificial intelligence can make healthcare human again. Basic Books, 2019.</mixed-citation>
				<element-citation publication-type="book">
					<person-group person-group-type="author">
						<name>
							<surname>Topol</surname>
							<given-names>E</given-names>
						</name>
					</person-group>
					<source>Deep medicine: How artificial intelligence can make healthcare human again</source>
					<publisher-name>Basic Books</publisher-name>
					<year>2019</year>
				</element-citation>
			</ref>
		</ref-list>
		<fn-group>
			<fn fn-type="other" id="fn1">
				<label>1</label>
				<p>Miembro Titular de la Sociedad Argentina de Cardiología</p>
			</fn>
			<fn fn-type="other" id="fn3">
				<label>Consideraciones éticas</label>
				<p> No aplica</p>
			</fn>
		</fn-group>
	</back>
	<!--<sub-article article-type="reply" id="s1" xml:lang="es">
		<front-stub>
			<article-categories>
				<subj-group subj-group-type="heading">
					<subject>Articles</subject>
				</subj-group>
			</article-categories>
			<title-group>
				<article-title>Respuesta de los autores</article-title>
			</title-group>
			<contrib-group>
				<contrib contrib-type="author">
					<name>
						<surname>Di Gennaro</surname>
						<given-names>Federico</given-names>
					</name>
				</contrib>
			</contrib-group>
		</front-stub>
		<body>
			<p>Agradecemos al Dr. Luciano Battioni por sus precisos y enriquecedores comentarios sobre nuestro trabajo titulado “Capacidad predictiva de eventos en pacientes con hipertensión arterial mediante el análisis con redes neuronales artificiales del monitoreo ambulatorio de presión arterial en comparación con la estratificación de riesgo clínica”. Coincidimos plenamente en que, si bien los modelos clásicos de estratificación de riesgo son utilizados en la práctica clínica cotidiana, la incorporación de nuevas herramientas metodológicas como el análisis mediante redes neuronales artificiales representa una oportunidad que nos permitiría optimizar la precisión diagnóstica y pronóstica de diferentes variables como las descriptas en el presente estudio. </p>
			<p>Estas tecnologías permiten integrar una gran cantidad de datos de manera simultánea, identificar patrones y generar predicciones más precisas en comparación con herramientas de análisis metodológico que utilizamos habitualmente. </p>
			<p>Reconocemos, tal como señala el Dr. Battioni, que estos modelos no están exentos de limitaciones, como el riesgo de sobreajuste y la necesidad de validación externa. Sin embargo, consideramos que su desarrollo e implementación, cuidadosamente evaluados, pueden complementar nuestro análisis clínico, actuando como una valiosa herramienta de apoyo para tomar decisiones más precisas. </p>
			<p>Esperamos que este trabajo contribuya a fomentar el diálogo y la investigación interdisciplinaria entre la medicina clínica y las ciencias de datos, y agradecemos una vez más la atenta lectura y los valiosos aportes realizados en su carta.</p>
			<p>Cordialmente,</p>
			<sig-block>
				<sig>Federico Di Gennaro</sig>
			</sig-block>
		</body>
	</sub-article>
	<sub-article article-type="translation" id="s2" xml:lang="en">
		<front-stub>
			<article-categories>
				<subj-group subj-group-type="heading">
					<subject>LETTERS FROM READERS</subject>
				</subj-group>
			</article-categories>
			<title-group>
				<article-title>&quot;Doctor, Should I Be Worried About This?&quot;</article-title>
			</title-group>
			<contrib-group>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0001-5130-626X</contrib-id>
					<name>
						<surname>BATTIONI</surname>
						<given-names>LUCIANO</given-names>
					</name>
					<xref ref-type="aff" rid="aff2"><sup>1</sup></xref>
					<xref ref-type="fn" rid="fn1"><sup>MTSAC</sup></xref>
				</contrib>
				<aff id="aff2">
					<label>1</label>
					<institution content-type="original">Postgraduate Director in “Artificial Intelligence Applied to Health Sciences”. Universidad Nacional del Litoral. Santa Fe, Santa Fe</institution>
					<institution content-type="normalized">Universidad Nacional del Litoral</institution>
					<institution content-type="orgname">Universidad Nacional del Litoral</institution>
					<addr-line>
						<city>Santa Fe</city>
						<state>Santa Fe</state>
					</addr-line>
					<country country="AR">Argentina</country>
				</aff>
			</contrib-group>
			<author-notes>
				<corresp id="c2">
					<label>Correspondence</label>: Luciano Battioni, Calle 15 N.° 669<italic>,</italic> Mercedes, Buenos Aires. E-mail: <email>lucianobattioni@gmail.com</email>
				</corresp>
				<fn fn-type="conflict" id="fn4">
					<p>Conflicts of interest None declared. (See authors' conflict of interests forms on the web).</p>
				</fn>
			</author-notes>
		</front-stub>
		<body>
			<p>If we could differentiate the universal goals of the medical sciences, we might conclude that there are three: diagnosis, prognosis, and treatment.</p>
			<p>Their relative importance differs between patients and doctors. The question I have chosen to title this letter is probably the most important to the patient. However, it is the one that has received the least scientific and technical development.</p>
			<p>In daily practice we use prognostic tools consistently and even dogmatically. In fact, many times we try to use scores generated to predict an event X in one population and extrapolate them to an event Y in another. (<xref ref-type="bibr" rid="B6">1</xref>) Most of these tools have areas under the ROC curve ranging from 0.60 to 0.85. (<xref ref-type="bibr" rid="B7">2</xref>,<xref ref-type="bibr" rid="B8">3</xref>) If we offered someone these tools to detect fraudulent banking transactions, they would quickly shake our hand and show us the way out. </p>
			<p>This poor current predictive performance is due not only to multiple limitations and difficulties associated with healthcare data management, but also to the tools that have been used so far. In the work entitled <italic>Events Prediction Ability in Patients with Hypertension using Artificial Neural Network Analysis of Ambulatory Blood Pressure Monitoring Compared to Clinical Risk Stratification,</italic> Di Gennaro et al. developed a simple neural network model capable of predicting what will happen to our patients more accurately. (<xref ref-type="bibr" rid="B9">4</xref>)</p>
			<p>Beyond the limitations acknowledged by the authors, it is worth highlighting what this work represents: the introduction of Artificial Intelligence (AI) tools into clinical practice. The integration of AI into medicine will change our practice in ways we cannot yet imagine. By integrating multiple variables, creating ones that we did not know existed or linking facts that elude human analysis, we will be able to provide precision medicine. (<xref ref-type="bibr" rid="B10">5</xref>)</p>
			<p>But not all that glitters is gold. For example, neural networks tend to overfit, i.e. they have high internal validity, but when validated in external cohorts their performance can drop significantly.</p>
			<p>To conclude, this work represents one of the first instances of using AI tools in medicine at a national level and, despite its design limitations, it gives us a very small sample of what this integration could represent and encourages us to continue research in this area.</p>
		</body>
		<back>
			<ref-list>
				<title>REFERENCES</title>
				<ref id="B6">
					<label>1</label>
					<mixed-citation>1. Wu JT, Wang SL, Chu YJ, Long DY, Dong JZ, Fan XW, et al. CHADS2 and CHA2DS2-VASc Scores Predict the Risk of Ischemic Stroke Outcome in Patients with Interatrial Block without Atrial Fibrillation. J Atheroscler Thromb 2017;24:176-84. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.5551/jat.34900">https://doi.org/10.5551/jat.34900</ext-link>
					</mixed-citation>
					<element-citation publication-type="journal">
						<person-group person-group-type="author">
							<name>
								<surname>Wu</surname>
								<given-names>JT</given-names>
							</name>
							<name>
								<surname>Wang</surname>
								<given-names>SL</given-names>
							</name>
							<name>
								<surname>Chu</surname>
								<given-names>YJ</given-names>
							</name>
							<name>
								<surname>Long</surname>
								<given-names>DY</given-names>
							</name>
							<name>
								<surname>Dong</surname>
								<given-names>JZ</given-names>
							</name>
							<name>
								<surname>Fan</surname>
								<given-names>XW</given-names>
							</name>
						</person-group>
						<article-title>CHADS2 and CHA2DS2-VASc Scores Predict the Risk of Ischemic Stroke Outcome in Patients with Interatrial Block without Atrial Fibrillation</article-title>
						<source>J Atheroscler Thromb</source>
						<year>2017</year>
						<volume>24</volume>
						<fpage>176</fpage>
						<lpage>184</lpage>
						<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.5551/jat.34900">https://doi.org/10.5551/jat.34900</ext-link>
					</element-citation>
				</ref>
				<ref id="B7">
					<label>2</label>
					<mixed-citation>2. D&amp;apos;Agostino RB Sr, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008;117:743-53. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1161/CIRCULATIONAHA.107.699579">https://doi.org/10.1161/CIRCULATIONAHA.107.699579</ext-link>
					</mixed-citation>
					<element-citation publication-type="journal">
						<person-group person-group-type="author">
							<name>
								<surname>D&amp;apos;Agostino</surname>
								<given-names>RB</given-names>
								<suffix>Sr</suffix>
							</name>
							<name>
								<surname>Vasan</surname>
								<given-names>RS</given-names>
							</name>
							<name>
								<surname>Pencina</surname>
								<given-names>MJ</given-names>
							</name>
							<name>
								<surname>Wolf</surname>
								<given-names>PA</given-names>
							</name>
							<name>
								<surname>Cobain</surname>
								<given-names>M</given-names>
							</name>
							<name>
								<surname>Massaro</surname>
								<given-names>JM</given-names>
							</name>
						</person-group>
						<article-title>General cardiovascular risk profile for use in primary care the Framingham Heart Study</article-title>
						<source>Circulation</source>
						<year>2008</year>
						<volume>117</volume>
						<fpage>743</fpage>
						<lpage>753</lpage>
						<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1161/CIRCULATIONAHA.107.699579">https://doi.org/10.1161/CIRCULATIONAHA.107.699579</ext-link>
					</element-citation>
				</ref>
				<ref id="B8">
					<label>3</label>
					<mixed-citation>3 . Olesen JB, Lip GY, Hansen ML, Hansen PR, Tolstrup JS, Lindhardsen J, et al. Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study. BMJ 2011; 342:d124. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1136/bmj.d124">https://doi.org/10.1136/bmj.d124</ext-link>
					</mixed-citation>
					<element-citation publication-type="journal">
						<person-group person-group-type="author">
							<name>
								<surname>Olesen</surname>
								<given-names>JB</given-names>
							</name>
							<name>
								<surname>Lip</surname>
								<given-names>GY</given-names>
							</name>
							<name>
								<surname>Hansen</surname>
								<given-names>ML</given-names>
							</name>
							<name>
								<surname>Hansen</surname>
								<given-names>PR</given-names>
							</name>
							<name>
								<surname>Tolstrup</surname>
								<given-names>JS</given-names>
							</name>
							<name>
								<surname>Lindhardsen</surname>
								<given-names>J</given-names>
							</name>
							<etal/>
						</person-group>
						<article-title>Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study</article-title>
						<source>BMJ</source>
						<year>2011</year>
						<volume>342</volume>
						<fpage>d124</fpage>
						<lpage>d124</lpage>
						<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1136/bmj.d124">https://doi.org/10.1136/bmj.d124</ext-link>
					</element-citation>
				</ref>
				<ref id="B9">
					<label>4</label>
					<mixed-citation>4 . Di Gennaro F P, Catalano MP, Aguirre AG, Fernández ML, Llanos R, Pérez Lloretet S, et al. Events Prediction Ability in Patients with Hypertension using Artificial Neural Network Analysis of Ambulatory Blood Pressure Monitoring Compared to Clinical Risk Stratification. . Rev Argent Cardiol 2025;93:33-42. <ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7775/rac.v93.i1.20854">https://doi.org/10.7775/rac.v93.i1.20854</ext-link>
					</mixed-citation>
					<element-citation publication-type="journal">
						<person-group person-group-type="author">
							<name>
								<surname>Di Gennaro F</surname>
								<given-names>P</given-names>
							</name>
							<name>
								<surname>Catalano</surname>
								<given-names>MP</given-names>
							</name>
							<name>
								<surname>Aguirre</surname>
								<given-names>AG</given-names>
							</name>
							<name>
								<surname>Fernández</surname>
								<given-names>ML</given-names>
							</name>
							<name>
								<surname>Llanos</surname>
								<given-names>R</given-names>
							</name>
							<name>
								<surname>Pérez Lloretet</surname>
								<given-names>S</given-names>
							</name>
							<etal/>
						</person-group>
						<article-title>Events Prediction Ability in Patients with Hypertension using Artificial Neural Network Analysis of Ambulatory Blood Pressure Monitoring Compared to Clinical Risk Stratification.</article-title>
						<source>Rev Argent Cardiol</source>
						<year>2025</year>
						<volume>93</volume>
						<fpage>33</fpage>
						<lpage>42</lpage>
						<ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7775/rac.v93.i1.20854">https://doi.org/10.7775/rac.v93.i1.20854</ext-link>
					</element-citation>
				</ref>
				<ref id="B10">
					<label>5</label>
					<mixed-citation>5 . Topol E. Deep medicine: How artificial intelligence can make healthcare human again. Basic Books, 2019.</mixed-citation>
					<element-citation publication-type="book">
						<person-group person-group-type="author">
							<name>
								<surname>Topol</surname>
								<given-names>E</given-names>
							</name>
						</person-group>
						<source>Deep medicine: How artificial intelligence can make healthcare human again</source>
						<publisher-name>Basic Books</publisher-name>
						<year>2019</year>
					</element-citation>
				</ref>
			</ref-list>
			<fn-group>
				<fn fn-type="other" id="fn5">
					<label>5</label>
					<p>Ethical considerations Not applicable.</p>
				</fn>
			</fn-group>
		</back>
		<sub-article article-type="reply" id="s3" xml:lang="en">
			<front-stub>
				<article-categories>
					<subj-group subj-group-type="heading">
						<subject>Articles</subject>
					</subj-group>
				</article-categories>
				<title-group>
					<article-title>AUTHORS’ REPLY</article-title>
				</title-group>
				<contrib-group>
					<contrib contrib-type="author">
						<name>
							<surname>Di Gennaro</surname>
							<given-names>Federico</given-names>
						</name>
					</contrib>
				</contrib-group>
			</front-stub>
			<body>
				<p>We would like to thank Dr. Luciano Battioni for his accurate and enriching comments on our work entitled <italic>Events Prediction Ability in Patients with Hypertension using Artificial Neural Network Analysis of Ambulatory Blood Pressure Monitoring Compared to Clinical Risk Stratification.</italic> We agree that, while classical risk stratification models are used in the development of risk stratification models, the incorporation of new methodological tools, such as the analysis using artificial neural networks represents an opportunity that would allow us to optimize the diagnostic and prognostic accuracy of different variables, such as those described in this study. These technologies make it possible to simultaneously integrate a large amount of data, identify patterns and generate more accurate predictions compared to the methodological analysis tools that we usually use. </p>
				<p>We recognize, as Dr. Battioni points out, that these models are not exempt from limitations, such as the risk of overfitting and the need for external validation. However, we believe that their development and implementation, carefully evaluated, can complement our clinical analysis and act as a valuable supportive tool to make more accurate decisions. </p>
				<p>We hope that this work will contribute to the promotion of dialogue and interdisciplinary research between clinical medicine and data science, and we thank you once again for the careful reading and the valuable contributions you have made in your letter.</p>
				<p>Yours sincerely,</p>
				<sig-block>
					<sig>Federico Di Gennaro</sig>
				</sig-block>
			</body>
		</sub-article>
	</sub-article>-->
</article>